From owner-pgsql-hackers@hub.org Mon Mar 22 18:43:41 1999 Received: from renoir.op.net (root@renoir.op.net [209.152.193.4]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id SAA23978 for ; Mon, 22 Mar 1999 18:43:39 -0500 (EST) Received: from hub.org (majordom@hub.org [209.47.145.100]) by renoir.op.net (o1/$ Revision: 1.18 $) with ESMTP id SAA06472 for ; Mon, 22 Mar 1999 18:36:44 -0500 (EST) Received: from localhost (majordom@localhost) by hub.org (8.9.2/8.9.1) with SMTP id SAA92604; Mon, 22 Mar 1999 18:34:23 -0500 (EST) (envelope-from owner-pgsql-hackers@hub.org) Received: by hub.org (TLB v0.10a (1.23 tibbs 1997/01/09 00:29:32)); Mon, 22 Mar 1999 18:33:50 +0000 (EST) Received: (from majordom@localhost) by hub.org (8.9.2/8.9.1) id SAA92469 for pgsql-hackers-outgoing; Mon, 22 Mar 1999 18:33:47 -0500 (EST) (envelope-from owner-pgsql-hackers@postgreSQL.org) Received: from po8.andrew.cmu.edu (PO8.ANDREW.CMU.EDU [128.2.10.108]) by hub.org (8.9.2/8.9.1) with ESMTP id SAA92456 for ; Mon, 22 Mar 1999 18:33:41 -0500 (EST) (envelope-from er1p+@andrew.cmu.edu) Received: (from postman@localhost) by po8.andrew.cmu.edu (8.8.5/8.8.2) id SAA12894 for pgsql-hackers@postgresql.org; Mon, 22 Mar 1999 18:33:38 -0500 (EST) Received: via switchmail; Mon, 22 Mar 1999 18:33:38 -0500 (EST) Received: from cloudy.me.cmu.edu via qmail ID ; Mon, 22 Mar 1999 18:27:20 -0500 (EST) Received: from cloudy.me.cmu.edu via qmail ID ; Mon, 22 Mar 1999 18:27:17 -0500 (EST) Received: from mms.4.60.Jun.27.1996.03.05.56.sun4.41.EzMail.2.0.CUILIB.3.45.SNAP.NOT.LINKED.cloudy.me.cmu.edu.sun4m.412 via MS.5.6.cloudy.me.cmu.edu.sun4_41; Mon, 22 Mar 1999 18:27:15 -0500 (EST) Message-ID: Date: Mon, 22 Mar 1999 18:27:15 -0500 (EST) From: Erik Riedel To: pgsql-hackers@postgreSQL.org Subject: [HACKERS] optimizer and type question Sender: owner-pgsql-hackers@postgreSQL.org Precedence: bulk Status: RO [last week aggregation, this week, the optimizer] I have a somewhat general optimizer question/problem that I would like to get some input on - i.e. I'd like to know what is "supposed" to work here and what I should be expecting. Sadly, I think the patch for this is more involved than my last message. Using my favorite table these days: Table = lineitem +------------------------+----------------------------------+-------+ | Field | Type | Length| +------------------------+----------------------------------+-------+ | l_orderkey | int4 not null | 4 | | l_partkey | int4 not null | 4 | | l_suppkey | int4 not null | 4 | | l_linenumber | int4 not null | 4 | | l_quantity | float4 not null | 4 | | l_extendedprice | float4 not null | 4 | | l_discount | float4 not null | 4 | | l_tax | float4 not null | 4 | | l_returnflag | char() not null | 1 | | l_linestatus | char() not null | 1 | | l_shipdate | date | 4 | | l_commitdate | date | 4 | | l_receiptdate | date | 4 | | l_shipinstruct | char() not null | 25 | | l_shipmode | char() not null | 10 | | l_comment | char() not null | 44 | +------------------------+----------------------------------+-------+ Index: lineitem_index_ and the query: -- -- Query 1 -- explain select l_returnflag, l_linestatus, sum(l_quantity) as sum_qty, sum(l_extendedprice) as sum_base_price, sum(l_extendedprice*(1-l_discount)) as sum_disc_price, sum(l_extendedprice*(1-l_discount)*(1+l_tax)) as sum_charge, avg(l_quantity) as avg_qty, avg(l_extendedprice) as avg_price, avg(l_discount) as avg_disc, count(*) as count_order from lineitem where l_shipdate <= '1998-09-02'::date group by l_returnflag, l_linestatus order by l_returnflag, l_linestatus; note that I have eliminated the date calculation in my query of last week and manually replaced it with a constant (since this wasn't happening automatically - but let's not worry about that for now). And this is only an explain, we care about the optimizer. So we get: Sort (cost=34467.88 size=0 width=0) -> Aggregate (cost=34467.88 size=0 width=0) -> Group (cost=34467.88 size=0 width=0) -> Sort (cost=34467.88 size=0 width=0) -> Seq Scan on lineitem (cost=34467.88 size=200191 width=44) so let's think about the selectivity that is being chosen for the seq scan (the where l_shipdate <= '1998-09-02'). Turns out the optimizer is choosing "33%", even though the real answer is somewhere in 90+% (that's how the query is designed). So, why does it do that? Turns out that selectivity in this case is determined via plancat::restriction_selectivity() which calls into functionOID = 103 (intltsel) for operatorOID = 1096 (date "<=") on relation OID = 18663 (my lineitem). This all follows because of the description of 1096 (date "<=") in pg_operator. Looking at local1_template1.bki.source near line 1754 shows: insert OID = 1096 ( "<=" PGUID 0 <...> date_le intltsel intltjoinsel ) where we see that indeed, it thinks "intltsel" is the right function to use for "oprrest" in the case of dates. Question 1 - is intltsel the right thing for selectivity on dates? Hope someone is still with me. So now we're running selfuncs::intltsel() where we make a further call to selfuncs::gethilokey(). The job of gethilokey is to determine the min and max values of a particular attribute in the table, which will then be used with the constant in my where clause to estimate the selectivity. It is going to search the pg_statistic relation with three key values: Anum_pg_statistic_starelid 18663 (lineitem) Anum_pg_statistic_staattnum 11 (l_shipdate) Anum_pg_statistic_staop 1096 (date "<=") this finds no tuples in pg_statistic. Why is that? The only nearby tuple in pg_statistic is: starelid|staattnum|staop|stalokey |stahikey --------+---------+-----+----------------+---------------- 18663| 11| 0|01-02-1992 |12-01-1998 and the reason the query doesn't match anything? Because 1096 != 0. But why is it 0 in pg_statistic? Statistics are determined near line 1844 in vacuum.c (assuming a 'vacuum analyze' run at some point) i = 0; values[i++] = (Datum) relid; /* 1 */ values[i++] = (Datum) attp->attnum; /* 2 */ ====> values[i++] = (Datum) InvalidOid; /* 3 */ fmgr_info(stats->outfunc, &out_function); out_string = <...min...> values[i++] = (Datum) fmgr(F_TEXTIN, out_string); pfree(out_string); out_string = <...max...> values[i++] = (Datum) fmgr(F_TEXTIN, out_string); pfree(out_string); stup = heap_formtuple(sd->rd_att, values, nulls); the "offending" line is setting the staop to InvalidOid (i.e. 0). Question 2 - is this right? Is the intent for 0 to serve as a "wildcard", or should it be inserting an entry for each operation individually? In the case of "wildcard" then gethilokey() should allow a match for Anum_pg_statistic_staop 0 instead of requiring the more restrictive 1096. In the current code, what happens next is gethilokey() returns "not found" and intltsel() returns the default 1/3 which I see in the resultant query plan (size = 200191 is 1/3 of the number of lineitem tuples). Question 3 - is there any inherent reason it couldn't get this right? The statistic is in the table 1992 to 1998, so the '1998-09-02' date should be 90-some% selectivity, a much better guess than 33%. Doesn't make a difference for this particular query, of course, because the seq scan must proceed anyhow, but it could easily affect other queries where selectivities matter (and it affects the modifications I am trying to test in the optimizer to be "smarter" about selectivities - my overall context is to understand/improve the behavior that the underlying storage system sees from queries like this). OK, so let's say we treat 0 as a "wildcard" and stop checking for 1096. Not we let gethilokey() return the two dates from the statistic table. The immediate next thing that intltsel() does, near lines 122 in selfuncs.c is call atol() on the strings from gethilokey(). And guess what it comes up with? low = 1 high = 12 because it calls atol() on '01-02-1992' and '12-01-1998'. This clearly isn't right, it should get some large integer that includes the year and day in the result. Then it should compare reasonably with my constant from the where clause and give a decent selectivity value. This leads to a re-visit of Question 1. Question 4 - should date "<=" use a dateltsel() function instead of intltsel() as oprrest? If anyone is still with me, could you tell me if this makes sense, or if there is some other location where the appropriate type conversion could take place so that intltsel() gets something reasonable when it does the atol() calls? Could someone also give me a sense for how far out-of-whack the whole current selectivity-handling structure is? It seems that most of the operators in pg_operator actually use intltsel() and would have type-specific problems like that described. Or is the problem in the way attribute values are stored in pg_statistic by vacuum analyze? Or is there another layer where type conversion belongs? Phew. Enough typing, hope someone can follow this and address at least some of the questions. Thanks. Erik Riedel Carnegie Mellon University www.cs.cmu.edu/~riedel From owner-pgsql-hackers@hub.org Mon Mar 22 20:31:11 1999 Received: from renoir.op.net (root@renoir.op.net [209.152.193.4]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id UAA00802 for ; Mon, 22 Mar 1999 20:31:09 -0500 (EST) Received: from hub.org (majordom@hub.org [209.47.145.100]) by renoir.op.net (o1/$ Revision: 1.18 $) with ESMTP id UAA13231 for ; Mon, 22 Mar 1999 20:15:20 -0500 (EST) Received: from localhost (majordom@localhost) by hub.org (8.9.2/8.9.1) with SMTP id UAA01981; Mon, 22 Mar 1999 20:14:04 -0500 (EST) (envelope-from owner-pgsql-hackers@hub.org) Received: by hub.org (TLB v0.10a (1.23 tibbs 1997/01/09 00:29:32)); Mon, 22 Mar 1999 20:13:32 +0000 (EST) Received: (from majordom@localhost) by hub.org (8.9.2/8.9.1) id UAA01835 for pgsql-hackers-outgoing; Mon, 22 Mar 1999 20:13:28 -0500 (EST) (envelope-from owner-pgsql-hackers@postgreSQL.org) Received: from sss.sss.pgh.pa.us (sss.pgh.pa.us [206.210.65.6]) by hub.org (8.9.2/8.9.1) with ESMTP id UAA01822 for ; Mon, 22 Mar 1999 20:13:21 -0500 (EST) (envelope-from tgl@sss.pgh.pa.us) Received: from sss.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.sss.pgh.pa.us (8.9.1/8.9.1) with ESMTP id UAA23294; Mon, 22 Mar 1999 20:12:43 -0500 (EST) To: Erik Riedel cc: pgsql-hackers@postgreSQL.org Subject: Re: [HACKERS] optimizer and type question In-reply-to: Your message of Mon, 22 Mar 1999 18:27:15 -0500 (EST) Date: Mon, 22 Mar 1999 20:12:43 -0500 Message-ID: <23292.922151563@sss.pgh.pa.us> From: Tom Lane Sender: owner-pgsql-hackers@postgreSQL.org Precedence: bulk Status: ROr Erik Riedel writes: > [ optimizer doesn't find relevant pg_statistic entry ] It's clearly a bug that the selectivity code is not finding this tuple. If your analysis is correct, then selectivity estimation has *never* worked properly, or at least not in recent memory :-(. Yipes. Bruce and I found a bunch of other problems in the optimizer recently, so it doesn't faze me to assume that this is broken too. > the "offending" line is setting the staop to InvalidOid (i.e. 0). > Question 2 - is this right? Is the intent for 0 to serve as a > "wildcard", My thought is that what the staop column ought to be is the OID of the comparison function that was used to determine the sort order of the column. Without a sort op the lowest and highest keys in the column are not well defined, so it makes no sense to assert "these are the lowest and highest values" without providing the sort op that determined that. (For sufficiently complex data types one could reasonably have multiple ordering operators. A crude example is sorting on "circumference" and "area" for polygons.) But typically the sort op will be the "<" operator for the column data type. So, the vacuum code is definitely broken --- it's not storing the sort op that it used. The code in gethilokey might be broken too, depending on how it is producing the operator it's trying to match against the tuple. For example, if the actual operator in the query is any of < <= > >= on int4, then int4lt ought to be used to probe the pg_statistic table. I'm not sure if we have adequate info in pg_operator or pg_type to let the optimizer code determine the right thing to probe with :-( > The immediate next thing that intltsel() does, near lines 122 > in selfuncs.c is call atol() on the strings from gethilokey(). And > guess what it comes up with? > low = 1 > high = 12 > because it calls atol() on '01-02-1992' and '12-01-1998'. This > clearly isn't right, it should get some large integer that includes > the year and day in the result. Then it should compare reasonably > with my constant from the where clause and give a decent selectivity > value. This leads to a re-visit of Question 1. > Question 4 - should date "<=" use a dateltsel() function instead of > intltsel() as oprrest? This is clearly busted as well. I'm not sure that creating dateltsel() is the right fix, however, because if you go down that path then every single datatype needs its own selectivity function; that's more than we need. What we really want here is to be able to map datatype values into some sort of numeric range so that we can compute what fraction of the low-key-to-high-key range is on each side of the probe value (the constant taken from the query). This general concept will apply to many scalar types, so what we want is a type-specific mapping function and a less-specific fraction-computing-function. Offhand I'd say that we want intltsel() and floatltsel(), plus conversion routines that can produce either int4 or float8 from a data type as seems appropriate. Anything that couldn't map to one or the other would have to supply its own selectivity function. > Or is the problem in the > way attribute values are stored in pg_statistic by vacuum analyze? Looks like it converts the low and high values to text and stores them that way. Ugly as can be :-( but I'm not sure there is a good alternative. We have no "wild card" column type AFAIK, which is what these columns of pg_statistic would have to be to allow storage of unconverted min and max values. I think you've found a can of worms here. Congratulations ;-) regards, tom lane From owner-pgsql-hackers@hub.org Mon Mar 22 23:31:00 1999 Received: from renoir.op.net (root@renoir.op.net [209.152.193.4]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id XAA03384 for ; Mon, 22 Mar 1999 23:30:58 -0500 (EST) Received: from hub.org (majordom@hub.org [209.47.145.100]) by renoir.op.net (o1/$ Revision: 1.18 $) with ESMTP id XAA25586 for ; Mon, 22 Mar 1999 23:18:25 -0500 (EST) Received: from localhost (majordom@localhost) by hub.org (8.9.2/8.9.1) with SMTP id XAA17955; Mon, 22 Mar 1999 23:17:24 -0500 (EST) (envelope-from owner-pgsql-hackers@hub.org) Received: by hub.org (TLB v0.10a (1.23 tibbs 1997/01/09 00:29:32)); Mon, 22 Mar 1999 23:16:49 +0000 (EST) Received: (from majordom@localhost) by hub.org (8.9.2/8.9.1) id XAA17764 for pgsql-hackers-outgoing; Mon, 22 Mar 1999 23:16:46 -0500 (EST) (envelope-from owner-pgsql-hackers@postgreSQL.org) Received: from po8.andrew.cmu.edu (PO8.ANDREW.CMU.EDU [128.2.10.108]) by hub.org (8.9.2/8.9.1) with ESMTP id XAA17745 for ; Mon, 22 Mar 1999 23:16:39 -0500 (EST) (envelope-from er1p+@andrew.cmu.edu) Received: (from postman@localhost) by po8.andrew.cmu.edu (8.8.5/8.8.2) id XAA04273; Mon, 22 Mar 1999 23:16:37 -0500 (EST) Received: via switchmail; Mon, 22 Mar 1999 23:16:37 -0500 (EST) Received: from hazy.adsl.net.cmu.edu via qmail ID ; Mon, 22 Mar 1999 23:15:09 -0500 (EST) Received: from hazy.adsl.net.cmu.edu via qmail ID ; Mon, 22 Mar 1999 23:15:00 -0500 (EST) Received: from mms.4.60.Jun.27.1996.03.02.53.sun4.51.EzMail.2.0.CUILIB.3.45.SNAP.NOT.LINKED.hazy.adsl.net.cmu.edu.sun4m.54 via MS.5.6.hazy.adsl.net.cmu.edu.sun4_51; Mon, 22 Mar 1999 23:14:55 -0500 (EST) Message-ID: <4qxlJ0200anI01hK40@andrew.cmu.edu> Date: Mon, 22 Mar 1999 23:14:55 -0500 (EST) From: Erik Riedel To: Tom Lane Subject: Re: [HACKERS] optimizer and type question Cc: pgsql-hackers@postgreSQL.org In-Reply-To: <23292.922151563@sss.pgh.pa.us> References: <23292.922151563@sss.pgh.pa.us> Sender: owner-pgsql-hackers@postgreSQL.org Precedence: bulk Status: ROr OK, building on your high-level explanation, I am attaching a patch that attempts to do something "better" than the current code. Note that I have only tested this with the date type and my particular query. I haven't run it through the regression, so consider it "proof of concept" at best. Although hopefully it will serve my purposes. > My thought is that what the staop column ought to be is the OID of the > comparison function that was used to determine the sort order of the > column. Without a sort op the lowest and highest keys in the column are > not well defined, so it makes no sense to assert "these are the lowest > and highest values" without providing the sort op that determined that. > > (For sufficiently complex data types one could reasonably have multiple > ordering operators. A crude example is sorting on "circumference" and > "area" for polygons.) But typically the sort op will be the "<" > operator for the column data type. > I changed vacuum.c to do exactly that. oid of the lt sort op. > So, the vacuum code is definitely broken --- it's not storing the sort > op that it used. The code in gethilokey might be broken too, depending > on how it is producing the operator it's trying to match against the > tuple. For example, if the actual operator in the query is any of > < <= > >= on int4, then int4lt ought to be used to probe the pg_statistic > table. I'm not sure if we have adequate info in pg_operator or pg_type > to let the optimizer code determine the right thing to probe with :-( > This indeed seems like a bigger problem. I thought about somehow using type-matching from the sort op and the actual operator in the query - if both the left and right type match, then consider them the same for purposes of this probe. That seemed complicated, so I punted in my example - it just does the search with relid and attnum and assumes that only returns one tuple. This works in my case (maybe in all cases, because of the way vacuum is currently written - ?). > What we really want here is to be able to map datatype values into > some sort of numeric range so that we can compute what fraction of the > low-key-to-high-key range is on each side of the probe value (the > constant taken from the query). This general concept will apply to > many scalar types, so what we want is a type-specific mapping function > and a less-specific fraction-computing-function. Offhand I'd say that > we want intltsel() and floatltsel(), plus conversion routines that can > produce either int4 or float8 from a data type as seems appropriate. > Anything that couldn't map to one or the other would have to supply its > own selectivity function. > This is what my example then does. Uses the stored sort op to get the type and then uses typinput to convert from the string to an int4. Then puts the int4 back into string format because that's what everyone was expecting. It seems to work for my particular query. I now get: (selfuncs) gethilokey() obj 18663 attr 11 opid 1096 (ignored) (selfuncs) gethilokey() found op 1087 in pg_proc (selfuncs) gethilokey() found type 1082 in pg_type (selfuncs) gethilokey() going to use 1084 to convert type 1082 (selfuncs) gethilokey() have low -2921 high -396 (selfuncs) intltsel() high -396 low -2921 val -486 (plancat) restriction_selectivity() for func 103 op 1096 rel 18663 attr 11 const -486 flag 3 returns 0.964356 NOTICE: QUERY PLAN: Sort (cost=34467.88 size=0 width=0) -> Aggregate (cost=34467.88 size=0 width=0) -> Group (cost=34467.88 size=0 width=0) -> Sort (cost=34467.88 size=0 width=0) -> Seq Scan on lineitem (cost=34467.88 size=579166 width=44) including my printfs, which exist in the patch as well. Selectivity is now the expected 96% and the size estimate for the seq scan is much closer to correct. Again, not tested with anything besides date, so caveat not-tested. Hope this helps. Erik ----------------------[optimizer_fix.sh]------------------------ #! /bin/sh # This is a shell archive, meaning: # 1. Remove everything above the #! /bin/sh line. # 2. Save the resulting text in a file. # 3. Execute the file with /bin/sh (not csh) to create: # selfuncs.c.diff # vacuum.c.diff # This archive created: Mon Mar 22 22:58:14 1999 export PATH; PATH=/bin:/usr/bin:$PATH if test -f 'selfuncs.c.diff' then echo shar: "will not over-write existing file 'selfuncs.c.diff'" else cat << \SHAR_EOF > 'selfuncs.c.diff' *** /afs/ece.cmu.edu/project/lcs/lcs-004/er1p/postgres/611/src/backend/utils/adt /selfuncs.c Thu Mar 11 23:59:35 1999 --- /afs/ece.cmu.edu/project/lcs/lcs-004/er1p/postgres/615/src/backend/utils/adt /selfuncs.c Mon Mar 22 22:57:25 1999 *************** *** 32,37 **** --- 32,40 ---- #include "utils/lsyscache.h" /* for get_oprrest() */ #include "catalog/pg_statistic.h" + #include "catalog/pg_proc.h" /* for Form_pg_proc */ + #include "catalog/pg_type.h" /* for Form_pg_type */ + /* N is not a valid var/constant or relation id */ #define NONVALUE(N) ((N) == -1) *************** *** 103,110 **** bottom; result = (float64) palloc(sizeof(float64data)); ! if (NONVALUE(attno) || NONVALUE(relid)) *result = 1.0 / 3; else { /* XXX val = atol(value); */ --- 106,114 ---- bottom; result = (float64) palloc(sizeof(float64data)); ! if (NONVALUE(attno) || NONVALUE(relid)) { *result = 1.0 / 3; + } else { /* XXX val = atol(value); */ *************** *** 117,130 **** } high = atol(highchar); low = atol(lowchar); if ((flag & SEL_RIGHT && val < low) || (!(flag & SEL_RIGHT) && val > high)) { float32data nvals; nvals = getattdispersion(relid, (int) attno); ! if (nvals == 0) *result = 1.0 / 3.0; else { *result = 3.0 * (float64data) nvals; --- 121,136 ---- } high = atol(highchar); low = atol(lowchar); + printf("(selfuncs) intltsel() high %d low %d val %d\n",high,low,val); if ((flag & SEL_RIGHT && val < low) || (!(flag & SEL_RIGHT) && val > high)) { float32data nvals; nvals = getattdispersion(relid, (int) attno); ! if (nvals == 0) { *result = 1.0 / 3.0; + } else { *result = 3.0 * (float64data) nvals; *************** *** 336,341 **** --- 342,353 ---- { Relation rel; HeapScanDesc scan; + /* this assumes there is only one row in the statistics table for any particular */ + /* relid, attnum pair - could be more complicated if staop is also used. */ + /* at the moment, if there are multiple rows, this code ends up picking the */ + /* "first" one - er1p */ + /* the actual "ignoring" is done in the call to heap_beginscan() below, where */ + /* we only mention 2 of the 3 keys in this array - er1p */ static ScanKeyData key[3] = { {0, Anum_pg_statistic_starelid, F_OIDEQ, {0, 0, F_OIDEQ}}, {0, Anum_pg_statistic_staattnum, F_INT2EQ, {0, 0, F_INT2EQ}}, *************** *** 344,355 **** bool isnull; HeapTuple tuple; rel = heap_openr(StatisticRelationName); key[0].sk_argument = ObjectIdGetDatum(relid); key[1].sk_argument = Int16GetDatum((int16) attnum); key[2].sk_argument = ObjectIdGetDatum(opid); ! scan = heap_beginscan(rel, 0, SnapshotNow, 3, key); tuple = heap_getnext(scan, 0); if (!HeapTupleIsValid(tuple)) { --- 356,377 ---- bool isnull; HeapTuple tuple; + HeapTuple tup; + Form_pg_proc proc; + Form_pg_type typ; + Oid which_op; + Oid which_type; + int32 low_value; + int32 high_value; + rel = heap_openr(StatisticRelationName); key[0].sk_argument = ObjectIdGetDatum(relid); key[1].sk_argument = Int16GetDatum((int16) attnum); key[2].sk_argument = ObjectIdGetDatum(opid); ! printf("(selfuncs) gethilokey() obj %d attr %d opid %d (ignored)\n", ! key[0].sk_argument,key[1].sk_argument,key[2].sk_argument); ! scan = heap_beginscan(rel, 0, SnapshotNow, 2, key); tuple = heap_getnext(scan, 0); if (!HeapTupleIsValid(tuple)) { *************** *** 376,383 **** --- 398,461 ---- &isnull)); if (isnull) elog(DEBUG, "gethilokey: low key is null"); + heap_endscan(scan); heap_close(rel); + + /* now we deal with type conversion issues */ + /* when intltsel() calls this routine (who knows what other callers might do) */ + /* it assumes that it can call atol() on the strings and then use integer */ + /* comparison from there. what we are going to do here, then, is try to use */ + /* the type information from Anum_pg_statistic_staop to convert the high */ + /* and low values - er1p */ + + /* WARNING: this code has only been tested with the date type and has NOT */ + /* been regression tested. consider it "sample" code of what might be the */ + /* right kind of thing to do - er1p */ + + /* get the 'op' from pg_statistic and look it up in pg_proc */ + which_op = heap_getattr(tuple, + Anum_pg_statistic_staop, + RelationGetDescr(rel), + &isnull); + if (InvalidOid == which_op) { + /* ignore all this stuff, try conversion only if we have a valid staop */ + /* note that there is an accompanying change to 'vacuum analyze' that */ + /* gets this set to something useful. */ + } else { + /* staop looks valid, so let's see what we can do about conversion */ + tup = SearchSysCacheTuple(PROOID, ObjectIdGetDatum(which_op), 0, 0, 0); + if (!HeapTupleIsValid(tup)) { + elog(ERROR, "selfuncs: unable to find op in pg_proc %d", which_op); + } + printf("(selfuncs) gethilokey() found op %d in pg_proc\n",which_op); + + /* use that to determine the type of stahikey and stalokey via pg_type */ + proc = (Form_pg_proc) GETSTRUCT(tup); + which_type = proc->proargtypes[0]; /* XXX - use left and right separately? */ + tup = SearchSysCacheTuple(TYPOID, ObjectIdGetDatum(which_type), 0, 0, 0); + if (!HeapTupleIsValid(tup)) { + elog(ERROR, "selfuncs: unable to find type in pg_type %d", which_type); + } + printf("(selfuncs) gethilokey() found type %d in pg_type\n",which_type); + + /* and use that type to get the conversion function to int4 */ + typ = (Form_pg_type) GETSTRUCT(tup); + printf("(selfuncs) gethilokey() going to use %d to convert type %d\n",typ->typinput,which_type); + + /* and convert the low and high strings */ + low_value = (int32) fmgr(typ->typinput, *low, -1); + high_value = (int32) fmgr(typ->typinput, *high, -1); + printf("(selfuncs) gethilokey() have low %d high %d\n",low_value,high_value); + + /* now we have int4's, which we put back into strings because that's what out */ + /* callers (intltsel() at least) expect - er1p */ + pfree(*low); pfree(*high); /* let's not leak the old strings */ + *low = int4out(low_value); + *high = int4out(high_value); + + /* XXX - this probably leaks the two tups we got from SearchSysCacheTuple() - er1p */ + } } float64 SHAR_EOF fi if test -f 'vacuum.c.diff' then echo shar: "will not over-write existing file 'vacuum.c.diff'" else cat << \SHAR_EOF > 'vacuum.c.diff' *** /afs/ece.cmu.edu/project/lcs/lcs-004/er1p/postgres/611/src/backend/commands/ vacuum.c Thu Mar 11 23:59:09 1999 --- /afs/ece.cmu.edu/project/lcs/lcs-004/er1p/postgres/615/src/backend/commands/ vacuum.c Mon Mar 22 21:23:15 1999 *************** *** 1842,1848 **** i = 0; values[i++] = (Datum) relid; /* 1 */ values[i++] = (Datum) attp->attnum; /* 2 */ ! values[i++] = (Datum) InvalidOid; /* 3 */ fmgr_info(stats->outfunc, &out_function); out_string = (*fmgr_faddr(&out_function)) (stats->min, stats->attr->atttypid); values[i++] = (Datum) fmgr(F_TEXTIN, out_string); --- 1842,1848 ---- i = 0; values[i++] = (Datum) relid; /* 1 */ values[i++] = (Datum) attp->attnum; /* 2 */ ! values[i++] = (Datum) stats->f_cmplt.fn_oid; /* 3 */ /* get the '<' oid, instead of 'invalid' - er1p */ fmgr_info(stats->outfunc, &out_function); out_string = (*fmgr_faddr(&out_function)) (stats->min, stats->attr->atttypid); values[i++] = (Datum) fmgr(F_TEXTIN, out_string); SHAR_EOF fi exit 0 # End of shell archive From owner-pgsql-hackers@hub.org Tue Mar 23 12:31:05 1999 Received: from renoir.op.net (root@renoir.op.net [209.152.193.4]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id MAA17491 for ; Tue, 23 Mar 1999 12:31:04 -0500 (EST) Received: from hub.org (majordom@hub.org [209.47.145.100]) by renoir.op.net (o1/$ Revision: 1.18 $) with ESMTP id MAA08839 for ; Tue, 23 Mar 1999 12:08:14 -0500 (EST) Received: from localhost (majordom@localhost) by hub.org (8.9.2/8.9.1) with SMTP id MAA93649; Tue, 23 Mar 1999 12:04:57 -0500 (EST) (envelope-from owner-pgsql-hackers@hub.org) Received: by hub.org (TLB v0.10a (1.23 tibbs 1997/01/09 00:29:32)); Tue, 23 Mar 1999 12:03:00 +0000 (EST) Received: (from majordom@localhost) by hub.org (8.9.2/8.9.1) id MAA93355 for pgsql-hackers-outgoing; Tue, 23 Mar 1999 12:02:55 -0500 (EST) (envelope-from owner-pgsql-hackers@postgreSQL.org) Received: from sss.sss.pgh.pa.us (sss.pgh.pa.us [206.210.65.6]) by hub.org (8.9.2/8.9.1) with ESMTP id MAA93336 for ; Tue, 23 Mar 1999 12:02:43 -0500 (EST) (envelope-from tgl@sss.pgh.pa.us) Received: from sss.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.sss.pgh.pa.us (8.9.1/8.9.1) with ESMTP id MAA24455; Tue, 23 Mar 1999 12:01:57 -0500 (EST) To: Erik Riedel cc: pgsql-hackers@postgreSQL.org Subject: Re: [HACKERS] optimizer and type question In-reply-to: Your message of Mon, 22 Mar 1999 23:14:55 -0500 (EST) <4qxlJ0200anI01hK40@andrew.cmu.edu> Date: Tue, 23 Mar 1999 12:01:57 -0500 Message-ID: <24453.922208517@sss.pgh.pa.us> From: Tom Lane Sender: owner-pgsql-hackers@postgreSQL.org Precedence: bulk Status: RO Erik Riedel writes: > OK, building on your high-level explanation, I am attaching a patch that > attempts to do something "better" than the current code. Note that I > have only tested this with the date type and my particular query. Glad to see you working on this. I don't like the details of your patch too much though ;-). Here are some suggestions for making it better. 1. I think just removing staop from the lookup in gethilokey is OK for now, though I'm dubious about Bruce's thought that we could delete that field entirely. As you observe, vacuum will not currently put more than one tuple for a column into pg_statistic, so we can just do the lookup with relid and attno and leave it at that. But I think we ought to leave the field there, with the idea that vacuum might someday compute more than one statistic for a data column. Fixing vacuum to put its sort op into the field is a good idea in the meantime. 2. The type conversion you're doing in gethilokey is a mess; I think what you ought to make it do is simply the inbound conversion of the string from pg_statistic into the internal representation for the column's datatype, and return that value as a Datum. It also needs a cleaner success/failure return convention --- this business with "n" return is ridiculously type-specific. Also, the best and easiest way to find the type to convert to is to look up the column type in the info for the given relid, not search pg_proc with the staop value. (I'm not sure that will even work, since there are pg_proc entries with wildcard argument types.) 3. The atol() calls currently found in intltsel are a type-specific cheat on what is conceptually a two-step process: * Convert the string stored in pg_statistic back to the internal form for the column data type. * Generate a numeric representation of the data value that can be used as an estimate of the range of values in the table. The second step is trivial for integers, which may obscure the fact that there are two steps involved, but nonetheless there are. If you think about applying selectivity logic to strings, say, it becomes clear that the second step is a necessary component of the process. Furthermore, the second step must also be applied to the probe value that's being passed into the selectivity operator. (The probe value is already in internal form, of course; but it is not necessarily in a useful numeric form.) We can do the first of these steps by applying the appropriate "XXXin" conversion function for the column data type, as you have done. The interesting question is how to do the second one. A really clean solution would require adding a column to pg_type that points to a function that will do the appropriate conversion. I'd be inclined to make all of these functions return "double" (float8) and just have one top-level selectivity routine for all data types that can use range-based selectivity logic. We could probably hack something together that would not use an explicit conversion function for each data type, but instead would rely on type-specific assumptions inside the selectivity routines. We'd need many more selectivity routines though (at least one for each of int, float4, float8, and text data types) so I'm not sure we'd really save any work compared to doing it right. BTW, now that I look at this issue it's real clear that the selectivity entries in pg_operator are horribly broken. The intltsel/intgtsel selectivity routines are currently applied to 32 distinct data types: regression=> select distinct typname,oprleft from pg_operator, pg_type regression-> where pg_type.oid = oprleft regression-> and oprrest in (103,104); typname |oprleft ---------+------- _aclitem | 1034 abstime | 702 bool | 16 box | 603 bpchar | 1042 char | 18 cidr | 650 circle | 718 date | 1082 datetime | 1184 float4 | 700 float8 | 701 inet | 869 int2 | 21 int4 | 23 int8 | 20 line | 628 lseg | 601 macaddr | 829 money | 790 name | 19 numeric | 1700 oid | 26 oid8 | 30 path | 602 point | 600 polygon | 604 text | 25 time | 1083 timespan | 1186 timestamp| 1296 varchar | 1043 (32 rows) many of which are very obviously not compatible with integer for *any* purpose. It looks to me like a lot of data types were added to pg_operator just by copy-and-paste, without paying attention to whether the selectivity routines were actually correct for the data type. As the code stands today, the bogus entries don't matter because gethilokey always fails, so we always get 1/3 as the selectivity estimate for any comparison operator (except = and != of course). I had actually noticed that fact and assumed that it was supposed to work that way :-(. But, clearly, there is code in here that is *trying* to be smarter. As soon as we fix gethilokey so that it can succeed, we will start getting essentially-random selectivity estimates for those data types that aren't actually binary-compatible with integer. That will not do; we have to do something about the issue. regards, tom lane From tgl@sss.pgh.pa.us Tue Mar 23 12:31:02 1999 Received: from renoir.op.net (root@renoir.op.net [209.152.193.4]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id MAA17484 for ; Tue, 23 Mar 1999 12:31:01 -0500 (EST) Received: from sss.sss.pgh.pa.us (sss.pgh.pa.us [206.210.65.6]) by renoir.op.net (o1/$ Revision: 1.18 $) with ESMTP id MAA09042 for ; Tue, 23 Mar 1999 12:10:55 -0500 (EST) Received: from sss.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.sss.pgh.pa.us (8.9.1/8.9.1) with ESMTP id MAA24474; Tue, 23 Mar 1999 12:09:52 -0500 (EST) To: Bruce Momjian cc: riedel+@CMU.EDU, pgsql-hackers@postgreSQL.org Subject: Re: [HACKERS] optimizer and type question In-reply-to: Your message of Mon, 22 Mar 1999 21:25:45 -0500 (EST) <199903230225.VAA01641@candle.pha.pa.us> Date: Tue, 23 Mar 1999 12:09:52 -0500 Message-ID: <24471.922208992@sss.pgh.pa.us> From: Tom Lane Status: RO Bruce Momjian writes: > What we really need is some way to determine how far the requested value > is from the min/max values. With int, we just do (val-min)/(max-min). > That works, but how do we do that for types that don't support division. > Strings come to mind in this case. What I'm envisioning is that we still apply the (val-min)/(max-min) logic, but apply it to numeric values that are produced in a type-dependent way. For ints and floats the conversion is trivial, of course. For strings, the first thing that comes to mind is to return 0 for a null string and the value of the first byte for a non-null string. This would give you one-part-in-256 selectivity which is plenty good enough for what the selectivity code needs to do. (Actually, it's only that good if the strings' first bytes are pretty well spread out. If you have a table containing English words, for example, you might only get about one part in 26 this way, since the first bytes will probably only run from A to Z. Might be better to use the first two characters of the string to compute the selectivity representation.) In general, you can apply this logic as long as you can come up with some numerical approximation to the data type's sorting order. It doesn't have to be exact. regards, tom lane From owner-pgsql-hackers@hub.org Tue Mar 23 12:31:03 1999 Received: from renoir.op.net (root@renoir.op.net [209.152.193.4]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id MAA17488 for ; Tue, 23 Mar 1999 12:31:02 -0500 (EST) Received: from hub.org (majordom@hub.org [209.47.145.100]) by renoir.op.net (o1/$ Revision: 1.18 $) with ESMTP id MAA09987 for ; Tue, 23 Mar 1999 12:21:34 -0500 (EST) Received: from localhost (majordom@localhost) by hub.org (8.9.2/8.9.1) with SMTP id MAA95155; Tue, 23 Mar 1999 12:18:33 -0500 (EST) (envelope-from owner-pgsql-hackers@hub.org) Received: by hub.org (TLB v0.10a (1.23 tibbs 1997/01/09 00:29:32)); Tue, 23 Mar 1999 12:17:00 +0000 (EST) Received: (from majordom@localhost) by hub.org (8.9.2/8.9.1) id MAA94857 for pgsql-hackers-outgoing; Tue, 23 Mar 1999 12:16:56 -0500 (EST) (envelope-from owner-pgsql-hackers@postgreSQL.org) Received: from sss.sss.pgh.pa.us (sss.pgh.pa.us [206.210.65.6]) by hub.org (8.9.2/8.9.1) with ESMTP id MAA94469 for ; Tue, 23 Mar 1999 12:11:33 -0500 (EST) (envelope-from tgl@sss.pgh.pa.us) Received: from sss.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.sss.pgh.pa.us (8.9.1/8.9.1) with ESMTP id MAA24474; Tue, 23 Mar 1999 12:09:52 -0500 (EST) To: Bruce Momjian cc: riedel+@CMU.EDU, pgsql-hackers@postgreSQL.org Subject: Re: [HACKERS] optimizer and type question In-reply-to: Your message of Mon, 22 Mar 1999 21:25:45 -0500 (EST) <199903230225.VAA01641@candle.pha.pa.us> Date: Tue, 23 Mar 1999 12:09:52 -0500 Message-ID: <24471.922208992@sss.pgh.pa.us> From: Tom Lane Sender: owner-pgsql-hackers@postgreSQL.org Precedence: bulk Status: RO Bruce Momjian writes: > What we really need is some way to determine how far the requested value > is from the min/max values. With int, we just do (val-min)/(max-min). > That works, but how do we do that for types that don't support division. > Strings come to mind in this case. What I'm envisioning is that we still apply the (val-min)/(max-min) logic, but apply it to numeric values that are produced in a type-dependent way. For ints and floats the conversion is trivial, of course. For strings, the first thing that comes to mind is to return 0 for a null string and the value of the first byte for a non-null string. This would give you one-part-in-256 selectivity which is plenty good enough for what the selectivity code needs to do. (Actually, it's only that good if the strings' first bytes are pretty well spread out. If you have a table containing English words, for example, you might only get about one part in 26 this way, since the first bytes will probably only run from A to Z. Might be better to use the first two characters of the string to compute the selectivity representation.) In general, you can apply this logic as long as you can come up with some numerical approximation to the data type's sorting order. It doesn't have to be exact. regards, tom lane From owner-pgsql-hackers@hub.org Thu Jul 1 20:39:19 1999 Received: from hub.org (hub.org [209.167.229.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id UAA15403 for ; Thu, 1 Jul 1999 20:39:18 -0400 (EDT) Received: from hub.org (hub.org [209.167.229.1]) by hub.org (8.9.3/8.9.3) with ESMTP id UAA45018; Thu, 1 Jul 1999 20:20:27 -0400 (EDT) (envelope-from owner-pgsql-hackers@hub.org) Received: by hub.org (TLB v0.10a (1.23 tibbs 1997/01/09 00:29:32)); Thu, 01 Jul 1999 20:15:30 +0000 (EDT) Received: (from majordom@localhost) by hub.org (8.9.3/8.9.3) id UAA44474 for pgsql-hackers-outgoing; Thu, 1 Jul 1999 20:15:28 -0400 (EDT) (envelope-from owner-pgsql-hackers@postgreSQL.org) X-Authentication-Warning: hub.org: majordom set sender to owner-pgsql-hackers@postgreSQL.org using -f Received: from sd.tpf.co.jp (sd.tpf.co.jp [210.161.239.34]) by hub.org (8.9.3/8.9.3) with ESMTP id UAA44058 for ; Thu, 1 Jul 1999 20:12:10 -0400 (EDT) (envelope-from Inoue@tpf.co.jp) Received: from cadzone ([126.0.1.40] (may be forged)) by sd.tpf.co.jp (2.5 Build 2640 (Berkeley 8.8.6)/8.8.4) with SMTP id JAA00279 for ; Fri, 02 Jul 1999 09:11:58 +0900 From: "Hiroshi Inoue" To: "pgsql-hackers" Subject: [HACKERS] Optimization FAQ ? Date: Fri, 2 Jul 1999 09:14:10 +0900 Message-ID: <000401bec41f$ce81dcc0$2801007e@cadzone.tpf.co.jp> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-2022-jp" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook 8.5, Build 4.71.2173.0 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.2106.4 Importance: Normal Sender: owner-pgsql-hackers@postgreSQL.org Precedence: bulk Status: RO Hello all, I got the following result. It's FAQ ? drop table int2t; create table int2t (id int2 primary key); explain select * from int2t where id=1; NOTICE: QUERY PLAN: Seq Scan on int2t (cost=43.00 rows=2 width=2) explain select * from int2t where id=1::int2; NOTICE: QUERY PLAN: Index Scan using int2t_pkey on int2t (cost=2.05 rows=2 width=2) explain select * from int2t where id='1'; NOTICE: QUERY PLAN: Index Scan using int2t_pkey on int2t (cost=2.05 rows=2 width=2) Right behavior ? Regards. Hiroshi Inoue Inoue@tpf.co.jp From owner-pgsql-hackers@hub.org Thu Jan 20 18:45:32 2000 Received: from renoir.op.net (root@renoir.op.net [207.29.195.4]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id TAA00672 for ; Thu, 20 Jan 2000 19:45:30 -0500 (EST) Received: from hub.org (hub.org [216.126.84.1]) by renoir.op.net (o1/$Revision: 1.20 $) with ESMTP id TAA01989 for ; Thu, 20 Jan 2000 19:39:15 -0500 (EST) Received: from localhost (majordom@localhost) by hub.org (8.9.3/8.9.3) with SMTP id TAA00957; Thu, 20 Jan 2000 19:35:19 -0500 (EST) (envelope-from owner-pgsql-hackers) Received: by hub.org (bulk_mailer v1.5); Thu, 20 Jan 2000 19:33:34 -0500 Received: (from majordom@localhost) by hub.org (8.9.3/8.9.3) id TAA00581 for pgsql-hackers-outgoing; Thu, 20 Jan 2000 19:32:37 -0500 (EST) (envelope-from owner-pgsql-hackers@postgreSQL.org) Received: from sss2.sss.pgh.pa.us (sss.pgh.pa.us [209.114.166.2]) by hub.org (8.9.3/8.9.3) with ESMTP id TAA98940 for ; Thu, 20 Jan 2000 19:31:49 -0500 (EST) (envelope-from tgl@sss.pgh.pa.us) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss2.sss.pgh.pa.us (8.9.3/8.9.3) with ESMTP id TAA25390 for ; Thu, 20 Jan 2000 19:31:32 -0500 (EST) To: pgsql-hackers@postgreSQL.org Subject: [HACKERS] Some notes on optimizer cost estimates Date: Thu, 20 Jan 2000 19:31:32 -0500 Message-ID: <25387.948414692@sss.pgh.pa.us> From: Tom Lane Sender: owner-pgsql-hackers@postgreSQL.org Status: OR I have been spending some time measuring actual runtimes for various sequential-scan and index-scan query plans, and have learned that the current Postgres optimizer's cost estimation equations are not very close to reality at all. Presently we estimate the cost of a sequential scan as Nblocks + CPU_PAGE_WEIGHT * Ntuples --- that is, the unit of cost is the time to read one disk page, and we have a "fudge factor" that relates CPU time per tuple to disk time per page. (The default CPU_PAGE_WEIGHT is 0.033, which is probably too high for modern hardware --- 0.01 seems like it might be a better default, at least for simple queries.) OK, it's a simplistic model, but not too unreasonable so far. The cost of an index scan is measured in these same terms as Nblocks + CPU_PAGE_WEIGHT * Ntuples + CPU_INDEX_PAGE_WEIGHT * Nindextuples Here Ntuples is the number of tuples selected by the index qual condition (typically, it's less than the total table size used in sequential-scan estimation). CPU_INDEX_PAGE_WEIGHT essentially estimates the cost of scanning an index tuple; by default it's 0.017 or half CPU_PAGE_WEIGHT. Nblocks is estimated as the index size plus an appropriate fraction of the main table size. There are two big problems with this: 1. Since main-table tuples are visited in index order, we'll be hopping around from page to page in the table. The current cost estimation method essentially assumes that the buffer cache plus OS disk cache will be 100% efficient --- we will never have to read the same page of the main table twice in a scan, due to having discarded it between references. This of course is unreasonably optimistic. Worst case is that we'd fetch a main-table page for each selected tuple, but in most cases that'd be unreasonably pessimistic. 2. The cost of a disk page fetch is estimated at 1.0 unit for both sequential and index scans. In reality, sequential access is *much* cheaper than the quasi-random accesses performed by an index scan. This is partly a matter of physical disk seeks, and partly a matter of benefitting (or not) from any read-ahead logic the OS may employ. As best I can measure on my hardware, the cost of a nonsequential disk read should be estimated at 4 to 5 times the cost of a sequential one --- I'm getting numbers like 2.2 msec per disk page for sequential scans, and as much as 11 msec per page for index scans. I don't know, however, if this ratio is similar enough on other platforms to be useful for cost estimating. We could make it a parameter like we do for CPU_PAGE_WEIGHT ... but you know and I know that no one ever bothers to adjust those numbers in the field ... The other effect that needs to be modeled, and currently is not, is the "hit rate" of buffer cache. Presumably, this is 100% for tables smaller than the cache and drops off as the table size increases --- but I have no particular thoughts on the form of the dependency. Does anyone have ideas here? The problem is complicated by the fact that we don't really know how big the cache is; we know the number of buffers Postgres has, but we have no idea how big a disk cache the kernel is keeping. As near as I can tell, finding a hit in the kernel disk cache is not a lot more expensive than having the page sitting in Postgres' own buffers --- certainly it's much much cheaper than a disk read. BTW, if you want to do some measurements of your own, try turning on PGOPTIONS="-d 2 -te". This will dump a lot of interesting numbers into the postmaster log, if your platform supports getrusage(). regards, tom lane ************ From owner-pgsql-hackers@hub.org Thu Jan 20 20:26:33 2000 Received: from hub.org (hub.org [216.126.84.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id VAA06630 for ; Thu, 20 Jan 2000 21:26:32 -0500 (EST) Received: from localhost (majordom@localhost) by hub.org (8.9.3/8.9.3) with SMTP id VAA35022; Thu, 20 Jan 2000 21:22:08 -0500 (EST) (envelope-from owner-pgsql-hackers) Received: by hub.org (bulk_mailer v1.5); Thu, 20 Jan 2000 21:20:35 -0500 Received: (from majordom@localhost) by hub.org (8.9.3/8.9.3) id VAA34569 for pgsql-hackers-outgoing; Thu, 20 Jan 2000 21:19:38 -0500 (EST) (envelope-from owner-pgsql-hackers@postgreSQL.org) Received: from hercules.cs.ucsb.edu (hercules.cs.ucsb.edu [128.111.41.30]) by hub.org (8.9.3/8.9.3) with ESMTP id VAA34534 for ; Thu, 20 Jan 2000 21:19:26 -0500 (EST) (envelope-from xun@cs.ucsb.edu) Received: from xp10-06.dialup.commserv.ucsb.edu (root@xp10-06.dialup.commserv.ucsb.edu [128.111.253.249]) by hercules.cs.ucsb.edu (8.8.6/8.8.6) with ESMTP id SAA04655 for ; Thu, 20 Jan 2000 18:19:22 -0800 (PST) Received: from xp10-06.dialup.commserv.ucsb.edu (xun@localhost) by xp10-06.dialup.commserv.ucsb.edu (8.9.3/8.9.3) with ESMTP id SAA22377 for ; Thu, 20 Jan 2000 18:19:40 -0800 Message-Id: <200001210219.SAA22377@xp10-06.dialup.commserv.ucsb.edu> To: pgsql-hackers@postgreSQL.org Reply-to: xun@cs.ucsb.edu Subject: Re. [HACKERS] Some notes on optimizer cost estimates Date: Thu, 20 Jan 2000 18:19:40 -0800 From: Xun Cheng Sender: owner-pgsql-hackers@postgreSQL.org Status: OR I'm very glad you bring up this cost estimate issue. Recent work in database research have argued a more detailed disk access cost model should be used for large queries especially joins. Traditional cost estimate only considers the number of disk pages accessed. However a more detailed model would consider three parameters: avg. seek, avg. latency and avg. page transfer. For old disk, typical values are SEEK=9.5 milliseconds, LATENCY=8.3 ms, TRANSFER=2.6ms. A sequential continuous reading of a table (assuming 1000 continuous pages) would cost (SEEK+LATENCY+1000*TRANFER=2617.8ms); while quasi-randomly reading 200 times with 2 continuous pages/time would cost (SEEK+200*LATENCY+400*TRANSFER=2700ms). Someone from IBM lab re-studied the traditional ad hoc join algorithms (nested, sort-merge, hash) using the detailed cost model and found some interesting results. >I have been spending some time measuring actual runtimes for various >sequential-scan and index-scan query plans, and have learned that the >current Postgres optimizer's cost estimation equations are not very >close to reality at all. One interesting question I'd like to ask is if this non-closeness really affects the optimal choice of postgresql's query optimizer. And to what degree the effects might be? My point is that if the optimizer estimated the cost for sequential-scan is 10 and the cost for index-scan is 20 while the actual costs are 10 vs. 40, it should be ok because the optimizer would still choose sequential-scan as it should. >1. Since main-table tuples are visited in index order, we'll be hopping >around from page to page in the table. I'm not sure about the implementation in postgresql. One thing you might be able to do is to first collect all must-read page addresses from the index scan and then order them before the actual ordered page fetching. It would at least avoid the same page being read twice (not entirely true depending on the context (like in join) and algo.) >The current cost estimation >method essentially assumes that the buffer cache plus OS disk cache will >be 100% efficient --- we will never have to read the same page of the >main table twice in a scan, due to having discarded it between >references. This of course is unreasonably optimistic. Worst case >is that we'd fetch a main-table page for each selected tuple, but in >most cases that'd be unreasonably pessimistic. This is actually the motivation that I asked before if postgresql has a raw disk facility. That way we have much control on this cache issue. Of course only if we can provide some algo. better than OS cache algo. (depending on the context, like large joins), a raw disk facility will be worthwhile (besides the recoverability). Actually I have another question for you guys which is somehow related to this cost estimation issue. You know the difference between OLTP and OLAP. My question is how you target postgresql on both kinds of applications or just OLTP. From what I know OLTP and OLAP would have a big difference in query characteristics and thus optimization difference. If postgresql is only targeted on OLTP, the above cost estimation issue might not be that important. However for OLAP, large tables and large queries are common and optimization would be difficult. xun ************ From owner-pgsql-hackers@hub.org Thu Jan 20 20:41:44 2000 Received: from hub.org (hub.org [216.126.84.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id VAA07020 for ; Thu, 20 Jan 2000 21:41:43 -0500 (EST) Received: from localhost (majordom@localhost) by hub.org (8.9.3/8.9.3) with SMTP id VAA40222; Thu, 20 Jan 2000 21:34:08 -0500 (EST) (envelope-from owner-pgsql-hackers) Received: by hub.org (bulk_mailer v1.5); Thu, 20 Jan 2000 21:32:35 -0500 Received: (from majordom@localhost) by hub.org (8.9.3/8.9.3) id VAA38388 for pgsql-hackers-outgoing; Thu, 20 Jan 2000 21:31:38 -0500 (EST) (envelope-from owner-pgsql-hackers@postgreSQL.org) Received: from sss2.sss.pgh.pa.us (sss.pgh.pa.us [209.114.166.2]) by hub.org (8.9.3/8.9.3) with ESMTP id VAA37422 for ; Thu, 20 Jan 2000 21:31:02 -0500 (EST) (envelope-from tgl@sss.pgh.pa.us) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss2.sss.pgh.pa.us (8.9.3/8.9.3) with ESMTP id VAA26761; Thu, 20 Jan 2000 21:30:41 -0500 (EST) To: "Hiroshi Inoue" cc: pgsql-hackers@postgreSQL.org Subject: Re: [HACKERS] Some notes on optimizer cost estimates In-reply-to: <000b01bf63b1$093cbd40$2801007e@tpf.co.jp> References: <000b01bf63b1$093cbd40$2801007e@tpf.co.jp> Comments: In-reply-to "Hiroshi Inoue" message dated "Fri, 21 Jan 2000 10:44:20 +0900" Date: Thu, 20 Jan 2000 21:30:41 -0500 Message-ID: <26758.948421841@sss.pgh.pa.us> From: Tom Lane Sender: owner-pgsql-hackers@postgreSQL.org Status: ORr "Hiroshi Inoue" writes: > I've wondered why we cound't analyze database without vacuum. > We couldn't run vacuum light-heartedly because it acquires an > exclusive lock for the target table. There is probably no real good reason, except backwards compatibility, why the ANALYZE function (obtaining pg_statistic data) is part of VACUUM at all --- it could just as easily be a separate command that would only use read access on the database. Bruce is thinking about restructuring VACUUM, so maybe now is a good time to think about splitting out the ANALYZE code too. > In addition,vacuum error occurs with analyze option in most > cases AFAIK. Still, with current sources? What's the error message? I fixed a problem with pg_statistic tuples getting too big... regards, tom lane ************ From tgl@sss.pgh.pa.us Thu Jan 20 21:10:28 2000 Received: from sss2.sss.pgh.pa.us (sss.pgh.pa.us [209.114.166.2]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id WAA08412 for ; Thu, 20 Jan 2000 22:10:26 -0500 (EST) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss2.sss.pgh.pa.us (8.9.3/8.9.3) with ESMTP id WAA27080; Thu, 20 Jan 2000 22:10:28 -0500 (EST) To: Bruce Momjian cc: Hiroshi Inoue , pgsql-hackers@postgresql.org Subject: Re: [HACKERS] Some notes on optimizer cost estimates In-reply-to: <200001210248.VAA07186@candle.pha.pa.us> References: <200001210248.VAA07186@candle.pha.pa.us> Comments: In-reply-to Bruce Momjian message dated "Thu, 20 Jan 2000 21:48:57 -0500" Date: Thu, 20 Jan 2000 22:10:28 -0500 Message-ID: <27077.948424228@sss.pgh.pa.us> From: Tom Lane Status: OR Bruce Momjian writes: > It is nice that ANALYZE is done during vacuum. I can't imagine why you > would want to do an analyze without adding a vacuum to it. I guess > that's why I made them the same command. Well, the main bad thing about ANALYZE being part of VACUUM is that it adds to the length of time that VACUUM is holding an exclusive lock on the table. I think it'd make more sense for it to be a separate command. I have also been thinking about how to make ANALYZE produce a more reliable estimate of the most common value. The three-element list that it keeps now is a good low-cost hack, but it really doesn't produce a trustworthy answer unless the MCV is pretty darn C (since it will never pick up on the MCV at all until there are at least two occurrences in three adjacent tuples). The only idea I've come up with is to use a larger list, which would be slower and take more memory. I think that'd be OK in a separate command, but I hesitate to do it inside VACUUM --- VACUUM has its own considerable memory requirements, and there's still the issue of not holding down an exclusive lock longer than you have to. regards, tom lane From Inoue@tpf.co.jp Thu Jan 20 21:08:32 2000 Received: from sd.tpf.co.jp (sd.tpf.co.jp [210.161.239.34]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id WAA08225 for ; Thu, 20 Jan 2000 22:08:29 -0500 (EST) Received: from cadzone ([126.0.1.40] (may be forged)) by sd.tpf.co.jp (2.5 Build 2640 (Berkeley 8.8.6)/8.8.4) with SMTP id MAA04148; Fri, 21 Jan 2000 12:08:30 +0900 From: "Hiroshi Inoue" To: "Bruce Momjian" , "Tom Lane" Cc: Subject: RE: [HACKERS] Some notes on optimizer cost estimates Date: Fri, 21 Jan 2000 12:14:10 +0900 Message-ID: <001301bf63bd$95cbe680$2801007e@tpf.co.jp> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook 8.5, Build 4.71.2173.0 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 In-Reply-To: <200001210248.VAA07186@candle.pha.pa.us> Importance: Normal Status: OR > -----Original Message----- > From: Bruce Momjian [mailto:pgman@candle.pha.pa.us] > > > "Hiroshi Inoue" writes: > > > I've wondered why we cound't analyze database without vacuum. > > > We couldn't run vacuum light-heartedly because it acquires an > > > exclusive lock for the target table. > > > > There is probably no real good reason, except backwards compatibility, > > why the ANALYZE function (obtaining pg_statistic data) is part of > > VACUUM at all --- it could just as easily be a separate command that > > would only use read access on the database. Bruce is thinking about > > restructuring VACUUM, so maybe now is a good time to think about > > splitting out the ANALYZE code too. > > I put it in vacuum because at the time I didn't know how to do such > things and vacuum already scanned the table. I just linked on the the > scan. Seemed like a good idea at the time. > > It is nice that ANALYZE is done during vacuum. I can't imagine why you > would want to do an analyze without adding a vacuum to it. I guess > that's why I made them the same command. > > If I made them separate commands, both would have to scan the table, > though the analyze could do it without the exclusive lock, which would > be good. > The functionality of VACUUM and ANALYZE is quite different. I don't prefer to charge VACUUM more than now about analyzing database. Probably looong lock,more aborts .... Various kind of analysis would be possible by splitting out ANALYZE. Regards. Hiroshi Inoue Inoue@tpf.co.jp From owner-pgsql-hackers@hub.org Fri Jan 21 11:01:59 2000 Received: from hub.org (hub.org [216.126.84.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id MAA07821 for ; Fri, 21 Jan 2000 12:01:57 -0500 (EST) Received: from localhost (majordom@localhost) by hub.org (8.9.3/8.9.3) with SMTP id LAA77357; Fri, 21 Jan 2000 11:52:25 -0500 (EST) (envelope-from owner-pgsql-hackers) Received: by hub.org (bulk_mailer v1.5); Fri, 21 Jan 2000 11:50:46 -0500 Received: (from majordom@localhost) by hub.org (8.9.3/8.9.3) id LAA76756 for pgsql-hackers-outgoing; Fri, 21 Jan 2000 11:49:50 -0500 (EST) (envelope-from owner-pgsql-hackers@postgreSQL.org) Received: from eclipse.pacifier.com (eclipse.pacifier.com [199.2.117.78]) by hub.org (8.9.3/8.9.3) with ESMTP id LAA76594 for ; Fri, 21 Jan 2000 11:49:01 -0500 (EST) (envelope-from dhogaza@pacifier.com) Received: from desktop (dsl-dhogaza.pacifier.net [216.65.147.68]) by eclipse.pacifier.com (8.9.3/8.9.3pop) with SMTP id IAA00225; Fri, 21 Jan 2000 08:47:26 -0800 (PST) Message-Id: <3.0.1.32.20000121081044.01036290@mail.pacifier.com> X-Sender: dhogaza@mail.pacifier.com X-Mailer: Windows Eudora Pro Version 3.0.1 (32) Date: Fri, 21 Jan 2000 08:10:44 -0800 To: xun@cs.ucsb.edu, pgsql-hackers@postgreSQL.org From: Don Baccus Subject: Re: Re. [HACKERS] Some notes on optimizer cost estimates In-Reply-To: <200001210219.SAA22377@xp10-06.dialup.commserv.ucsb.edu> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-pgsql-hackers@postgreSQL.org Status: OR At 06:19 PM 1/20/00 -0800, Xun Cheng wrote: >I'm very glad you bring up this cost estimate issue. >Recent work in database research have argued a more >detailed disk access cost model should be used for >large queries especially joins. >Traditional cost estimate only considers the number of >disk pages accessed. However a more detailed model >would consider three parameters: avg. seek, avg. latency >and avg. page transfer. For old disk, typical values are >SEEK=9.5 milliseconds, LATENCY=8.3 ms, TRANSFER=2.6ms. >A sequential continuous reading of a table (assuming >1000 continuous pages) would cost >(SEEK+LATENCY+1000*TRANFER=2617.8ms); while quasi-randomly >reading 200 times with 2 continuous pages/time would >cost (SEEK+200*LATENCY+400*TRANSFER=2700ms). >Someone from IBM lab re-studied the traditional >ad hoc join algorithms (nested, sort-merge, hash) using the detailed cost model >and found some interesting results. One complication when doing an index scan is that you are accessing two separate files (table and index), which can frequently be expected to cause an considerable increase in average seek time. Oracle and other commercial databases recommend spreading indices and tables over several spindles if at all possible in order to minimize this effect. I suspect it also helps their optimizer make decisions that are more consistently good for customers with the largest and most complex databases and queries, by making cost estimates more predictably reasonable. Still...this doesn't help with the question about the effect of the filesystem system cache. I wandered around the web for a little bit last night, and found one summary of a paper by Osterhout on the effect of the Solaris cache on a fileserver serving diskless workstations. There was reference to the hierarchy involved (i.e. the local workstation cache is faster than the fileserver's cache which has to be read via the network which in turn is faster than reading from the fileserver's disk). It appears the rule-of-thumb for the cache-hit ratio on reads, presumably based on measuring some internal Sun systems, used in their calculations was 80%. Just a datapoint to think about. There's also considerable operating system theory on paging systems that might be useful for thinking about trying to estimate the Postgres cache/hit ratio. Then again, maybe Postgres could just keep count of how many pages of a given table are in the cache at any given time? Or simply keep track of the current ratio of hits and misses? >>I have been spending some time measuring actual runtimes for various >>sequential-scan and index-scan query plans, and have learned that the >>current Postgres optimizer's cost estimation equations are not very >>close to reality at all. >One interesting question I'd like to ask is if this non-closeness >really affects the optimal choice of postgresql's query optimizer. >And to what degree the effects might be? My point is that >if the optimizer estimated the cost for sequential-scan is 10 and >the cost for index-scan is 20 while the actual costs are 10 vs. 40, >it should be ok because the optimizer would still choose sequential-scan >as it should. This is crucial, of course - if there are only two types of scans available, what ever heuristic is used only has to be accurate enough to pick the right one. Once the choice is made, it doesn't really matter (from the optimizer's POV) just how long it will actually take, the time will be spent and presumably it will be shorter than the alternative. How frequently will the optimizer choose wrongly if: 1. All of the tables and indices were in PG buffer cache or filesystem cache? (i.e. fixed access times for both types of scans) or 2. The table's so big that only a small fraction can reside in RAM during the scan and join, which means that the non-sequential disk access pattern of the indexed scan is much more expensive. Also, if you pick sequential scans more frequently based on a presumption that index scans are expensive due to increased average seek time, how often will this penalize the heavy-duty user that invests in extra drives and lots of RAM? ... >>The current cost estimation >>method essentially assumes that the buffer cache plus OS disk cache will >>be 100% efficient --- we will never have to read the same page of the >>main table twice in a scan, due to having discarded it between >>references. This of course is unreasonably optimistic. Worst case >>is that we'd fetch a main-table page for each selected tuple, but in >>most cases that'd be unreasonably pessimistic. > >This is actually the motivation that I asked before if postgresql >has a raw disk facility. That way we have much control on this cache >issue. Of course only if we can provide some algo. better than OS >cache algo. (depending on the context, like large joins), a raw disk >facility will be worthwhile (besides the recoverability). Postgres does have control over its buffer cache. The one thing that raw disk I/O would give you is control over where blocks are placed, meaning you could more accurately model the cost of retrieving them. So presumably the cache could be tuned to the allocation algorithm used to place various structures on the disk. I still wonder just how much gain you get by this approach. Compared, to, say simply spending $2,000 on a gigabyte of RAM. Heck, PCs even support a couple gigs of RAM now. >Actually I have another question for you guys which is somehow related >to this cost estimation issue. You know the difference between OLTP >and OLAP. My question is how you target postgresql on both kinds >of applications or just OLTP. From what I know OLTP and OLAP would >have a big difference in query characteristics and thus >optimization difference. If postgresql is only targeted on >OLTP, the above cost estimation issue might not be that >important. However for OLAP, large tables and large queries are >common and optimization would be difficult. - Don Baccus, Portland OR Nature photos, on-line guides, Pacific Northwest Rare Bird Alert Service and other goodies at http://donb.photo.net. ************ From pgsql-hackers-owner+M6019@hub.org Mon Aug 21 11:47:56 2000 Received: from hub.org (root@hub.org [216.126.84.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id LAA07289 for ; Mon, 21 Aug 2000 11:47:55 -0400 (EDT) Received: from hub.org (majordom@localhost [127.0.0.1]) by hub.org (8.10.1/8.10.1) with SMTP id e7LFlpT03383; Mon, 21 Aug 2000 11:47:51 -0400 (EDT) Received: from mail.fct.unl.pt (fct1.si.fct.unl.pt [193.136.120.1]) by hub.org (8.10.1/8.10.1) with SMTP id e7LFlaT03243 for ; Mon, 21 Aug 2000 11:47:37 -0400 (EDT) Received: (qmail 7416 invoked by alias); 21 Aug 2000 15:54:33 -0000 Received: (qmail 7410 invoked from network); 21 Aug 2000 15:54:32 -0000 Received: from eros.si.fct.unl.pt (193.136.120.112) by fct1.si.fct.unl.pt with SMTP; 21 Aug 2000 15:54:32 -0000 Date: Mon, 21 Aug 2000 16:48:08 +0100 (WEST) From: =?iso-8859-1?Q?Tiago_Ant=E3o?= X-Sender: tiago@eros.si.fct.unl.pt To: Tom Lane cc: pgsql-hackers@postgresql.org Subject: Re: [HACKERS] Optimisation deficiency: currval('seq')-->seq scan, constant-->index scan In-Reply-To: <1731.966868649@sss.pgh.pa.us> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Mailing-List: pgsql-hackers@postgresql.org Precedence: bulk Sender: pgsql-hackers-owner@hub.org Status: ORr On Mon, 21 Aug 2000, Tom Lane wrote: > > One thing it might be interesting (please tell me if you think > > otherwise) would be to improve pg with better statistical information, by > > using, for example, histograms. > > Yes, that's been on the todo list for a while. If it's ok and nobody is working on that, I'll look on that subject. I'll start by looking at the analize portion of vacuum. I'm thinking in using arrays for the histogram (I've never used the array data type of postgres). Should I use 7.0.2 or the cvs version? > Interesting article. We do most of what she talks about, but we don't > have anything like the ClusterRatio statistic. We need it --- that was > just being discussed a few days ago in another thread. Do you have any > reference on exactly how DB2 defines that stat? I don't remember seeing that information spefically. From what I've read I can speculate: 1. They have clusterratios for both indexes and the relation itself. 2. They might use an index even if there is no "order by" if the table has a low clusterratio: just to get the RIDs, then sort the RIDs and fetch. 3. One possible way to calculate this ratio: a) for tables SeqScan if tuple points to a next tuple on the same page then its "good" ratio = # good tuples / # all tuples b) for indexes (high speculation ratio here) foreach pointed RID in index if RID is in same page of next RID in index than mark as "good" I suspect that if a tuple size is big (relative to page size) than the cluster ratio is always low. A tuple might also be "good" if it pointed to the next page. Tiago From pgsql-hackers-owner+M6152@hub.org Wed Aug 23 13:00:33 2000 Received: from hub.org (root@hub.org [216.126.84.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id NAA10259 for ; Wed, 23 Aug 2000 13:00:33 -0400 (EDT) Received: from hub.org (majordom@localhost [127.0.0.1]) by hub.org (8.10.1/8.10.1) with SMTP id e7NGsPN83008; Wed, 23 Aug 2000 12:54:25 -0400 (EDT) Received: from mail.fct.unl.pt (fct1.si.fct.unl.pt [193.136.120.1]) by hub.org (8.10.1/8.10.1) with SMTP id e7NGniN81749 for ; Wed, 23 Aug 2000 12:49:44 -0400 (EDT) Received: (qmail 9869 invoked by alias); 23 Aug 2000 15:10:04 -0000 Received: (qmail 9860 invoked from network); 23 Aug 2000 15:10:04 -0000 Received: from eros.si.fct.unl.pt (193.136.120.112) by fct1.si.fct.unl.pt with SMTP; 23 Aug 2000 15:10:04 -0000 Date: Wed, 23 Aug 2000 16:03:42 +0100 (WEST) From: =?iso-8859-1?Q?Tiago_Ant=E3o?= X-Sender: tiago@eros.si.fct.unl.pt To: Tom Lane cc: Jules Bean , pgsql-hackers@postgresql.org Subject: Re: [HACKERS] Optimisation deficiency: currval('seq')-->seq scan, constant-->index scan In-Reply-To: <27971.967041030@sss.pgh.pa.us> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Mailing-List: pgsql-hackers@postgresql.org Precedence: bulk Sender: pgsql-hackers-owner@hub.org Status: ORr Hi! On Wed, 23 Aug 2000, Tom Lane wrote: > Yes, we know about that one. We have stats about the most common value > in a column, but no information about how the less-common values are > distributed. We definitely need stats about several top values not just > one, because this phenomenon of a badly skewed distribution is pretty > common. An end-biased histogram has stats on top values and also on the least frequent values. So if a there is a selection on a value that is well bellow average, the selectivity estimation will be more acurate. On some research papers I've read, it's refered that this is a better approach than equi-width histograms (which are said to be the "industry" standard). I not sure whether to use a table or a array attribute on pg_stat for the histogram, the problem is what could be expected from the size of the attribute (being a text). I'm very affraid of the cost of going through several tuples on a table (pg_histogram?) during the optimization phase. One other idea would be to only have better statistics for special attributes requested by the user... something like "analyze special table(column)". Best Regards, Tiago From pgsql-hackers-owner+M6160@hub.org Thu Aug 24 00:21:39 2000 Received: from hub.org (root@hub.org [216.126.84.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id AAA27662 for ; Thu, 24 Aug 2000 00:21:38 -0400 (EDT) Received: from hub.org (majordom@localhost [127.0.0.1]) by hub.org (8.10.1/8.10.1) with SMTP id e7O46w585951; Thu, 24 Aug 2000 00:06:58 -0400 (EDT) Received: from sss2.sss.pgh.pa.us (sss.pgh.pa.us [209.114.166.2]) by hub.org (8.10.1/8.10.1) with ESMTP id e7O3uv583775 for ; Wed, 23 Aug 2000 23:56:57 -0400 (EDT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss2.sss.pgh.pa.us (8.9.3/8.9.3) with ESMTP id XAA20973; Wed, 23 Aug 2000 23:56:35 -0400 (EDT) To: =?iso-8859-1?Q?Tiago_Ant=E3o?= cc: Jules Bean , pgsql-hackers@postgresql.org Subject: Re: [HACKERS] Optimisation deficiency: currval('seq')-->seq scan, constant-->index scan In-reply-to: References: Comments: In-reply-to =?iso-8859-1?Q?Tiago_Ant=E3o?= message dated "Wed, 23 Aug 2000 16:03:42 +0100" Date: Wed, 23 Aug 2000 23:56:35 -0400 Message-ID: <20970.967089395@sss.pgh.pa.us> From: Tom Lane X-Mailing-List: pgsql-hackers@postgresql.org Precedence: bulk Sender: pgsql-hackers-owner@hub.org Status: OR =?iso-8859-1?Q?Tiago_Ant=E3o?= writes: > One other idea would be to only have better statistics for special > attributes requested by the user... something like "analyze special > table(column)". This might actually fall out "for free" from the cheapest way of implementing the stats. We've talked before about scanning btree indexes directly to obtain data values in sorted order, which makes it very easy to find the most common values. If you do that, you get good stats for exactly those columns that the user has created indexes on. A tad indirect but I bet it'd be effective... regards, tom lane From pgsql-hackers-owner+M6165@hub.org Thu Aug 24 05:33:02 2000 Received: from hub.org (root@hub.org [216.126.84.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id FAA14309 for ; Thu, 24 Aug 2000 05:33:01 -0400 (EDT) Received: from hub.org (majordom@localhost [127.0.0.1]) by hub.org (8.10.1/8.10.1) with SMTP id e7O9X0584670; Thu, 24 Aug 2000 05:33:00 -0400 (EDT) Received: from athena.office.vi.net (office-gwb.fulham.vi.net [194.88.77.158]) by hub.org (8.10.1/8.10.1) with ESMTP id e7O9Ix581216 for ; Thu, 24 Aug 2000 05:19:03 -0400 (EDT) Received: from grommit.office.vi.net [192.168.1.200] (mail) by athena.office.vi.net with esmtp (Exim 3.12 #1 (Debian)) id 13Rt2Y-00073I-00; Thu, 24 Aug 2000 10:11:14 +0100 Received: from jules by grommit.office.vi.net with local (Exim 3.12 #1 (Debian)) id 13Rt2Y-0005GV-00; Thu, 24 Aug 2000 10:11:14 +0100 Date: Thu, 24 Aug 2000 10:11:14 +0100 From: Jules Bean To: Tom Lane Cc: Tiago Ant?o , pgsql-hackers@postgresql.org Subject: Re: [HACKERS] Optimisation deficiency: currval('seq')-->seq scan, constant-->index scan Message-ID: <20000824101113.N17510@grommit.office.vi.net> References: <1731.966868649@sss.pgh.pa.us> <20000823133418.F17510@grommit.office.vi.net> <27971.967041030@sss.pgh.pa.us> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <27971.967041030@sss.pgh.pa.us>; from tgl@sss.pgh.pa.us on Wed, Aug 23, 2000 at 10:30:30AM -0400 X-Mailing-List: pgsql-hackers@postgresql.org Precedence: bulk Sender: pgsql-hackers-owner@hub.org Status: OR On Wed, Aug 23, 2000 at 10:30:30AM -0400, Tom Lane wrote: > Jules Bean writes: > > I have in a table a 'category' column which takes a small number of > > (basically fixed) values. Here by 'small', I mean ~1000, while the > > table itself has ~10 000 000 rows. Some categories have many, many > > more rows than others. In particular, there's one category which hits > > over half the rows. Because of this (AIUI) postgresql assumes > > that the query > > select ... from thistable where category='something' > > is best served by a seqscan, even though there is an index on > > category. > > Yes, we know about that one. We have stats about the most common value > in a column, but no information about how the less-common values are > distributed. We definitely need stats about several top values not just > one, because this phenomenon of a badly skewed distribution is pretty > common. ISTM that that might be enough, in fact. If you have stats telling you that the most popular value is 'xyz', and that it constitutes 50% of the rows (i.e. 5 000 000) then you can conclude that, on average, other entries constitute a mere 5 000 000/999 ~~ 5000 entries, and it would be definitely be enough. (That's assuming you store the number of distinct values somewhere). > BTW, if your highly-popular value is actually a dummy value ('UNKNOWN' > or something like that), a fairly effective workaround is to replace the > dummy entries with NULL. The system does account for NULLs separately > from real values, so you'd then get stats based on the most common > non-dummy value. I can't really do that. Even if I could, the distribution is very skewed -- so the next most common makes up a very high proportion of what's left. I forget the figures exactly. Jules From pgsql-hackers-owner+M6154@hub.org Wed Aug 23 14:36:41 2000 Received: from hub.org (root@hub.org [216.126.84.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id NAA11076 for ; Wed, 23 Aug 2000 13:36:41 -0400 (EDT) Received: from hub.org (majordom@localhost [127.0.0.1]) by hub.org (8.10.1/8.10.1) with SMTP id e7NHTqN92431; Wed, 23 Aug 2000 13:29:52 -0400 (EDT) Received: from mail.fct.unl.pt (fct1.si.fct.unl.pt [193.136.120.1]) by hub.org (8.10.1/8.10.1) with SMTP id e7NHM1N90883 for ; Wed, 23 Aug 2000 13:22:01 -0400 (EDT) Received: (qmail 13816 invoked by alias); 23 Aug 2000 17:29:02 -0000 Received: (qmail 13807 invoked from network); 23 Aug 2000 17:29:02 -0000 Received: from eros.si.fct.unl.pt (193.136.120.112) by fct1.si.fct.unl.pt with SMTP; 23 Aug 2000 17:29:02 -0000 Date: Wed, 23 Aug 2000 18:22:40 +0100 (WEST) From: =?iso-8859-1?Q?Tiago_Ant=E3o?= X-Sender: tiago@eros.si.fct.unl.pt To: Tom Lane cc: =?iso-8859-1?Q?Tiago_Ant=E3o?= , PostgreSQL Hackers list Subject: Re: [HACKERS] analyze.c In-Reply-To: <28154.967041988@sss.pgh.pa.us> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Mailing-List: pgsql-hackers@postgresql.org Precedence: bulk Sender: pgsql-hackers-owner@hub.org Status: ORr On Wed, 23 Aug 2000, Tom Lane wrote: > > What's the big reason not to do that? I know that > > there is some code in analyze.c (like comparing) that uses other parts of > > pg, but that seems to be easily fixed. > > Are you proposing not to do any comparisons? It will be interesting to > see how you can compute a histogram without any idea of equality or > ordering. But if you want that, then you still need the function-call > manager as well as the type-specific comparison routines for every > datatype that you might be asked to operate on (don't forget > user-defined types here). I forgot user defined data types :-(, but regarding histograms I think the code can be made external (at least for testing purposes): 1. I was not suggesting not to do any comparisons, but I think the only comparison I need is equality, I don't need order as I don't need to calculate mins or maxs (I just need mins and maxes on frequencies, NOT on dat itself) to make a histogram. 2. The mapping to text guarantees that I have (PQgetvalue returns always char* and pg_statistics keeps a "text" anyway) a way of knowing about equality regardless of type. But at least anything relating to order has to be in. > > I'm leaning toward the implementation of end-biased histograms. There is > > an introductory reference in the IEEE Data Engineering Bulletin, september > > 1995 (available on microsoft research site). > > Sounds interesting. Can you give us an exact URL? http://www.research.microsoft.com/research/db/debull/default.htm BTW, you can get access to SIGMOD CDs with lots of goodies for a very low price (at least in 1999 it was a bargain), check out ACM membership for sigmod. I've been reading something about implementation of histograms, and, AFAIK, in practice histograms is just a cool name for no more than: 1. top ten with frequency for each 2. the same for top ten worse 3. average for the rest I'm writing code get this info (outside pg for now - for testing purposes). Best Regards, Tiago PS - again: I'm starting, so, some of my comments can be completly dumb. From pgsql-hackers-owner+M7514@hub.org Sun Oct 15 20:38:12 2000 Received: from hub.org (hub.org [216.126.84.1]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id TAA18459 for ; Sun, 15 Oct 2000 19:38:12 -0400 (EDT) Received: from hub.org.org (localhost [127.0.0.1]) by hub.org (8.10.1/8.10.1) with SMTP id e9FNaUR59496; Sun, 15 Oct 2000 19:36:30 -0400 (EDT) Received: from sss.pgh.pa.us (sss.pgh.pa.us [209.114.132.154]) by hub.org (8.10.1/8.10.1) with ESMTP id e9FNYuR58276 for ; Sun, 15 Oct 2000 19:34:56 -0400 (EDT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.11.0/8.11.0) with ESMTP id e9FNXaB06046; Sun, 15 Oct 2000 19:33:36 -0400 (EDT) To: Bruce Momjian cc: Jules Bean , Alfred Perlstein , pgsql-hackers@postgresql.org Subject: Re: [HACKERS] Performance on inserts In-reply-to: <200010152320.TAA17944@candle.pha.pa.us> References: <200010152320.TAA17944@candle.pha.pa.us> Comments: In-reply-to Bruce Momjian message dated "Sun, 15 Oct 2000 19:20:35 -0400" Date: Sun, 15 Oct 2000 19:33:36 -0400 Message-ID: <6043.971652816@sss.pgh.pa.us> From: Tom Lane X-Mailing-List: pgsql-hackers@postgresql.org Precedence: bulk Sender: pgsql-hackers-owner@hub.org Status: ORr Bruce Momjian writes: > However, assume tab2.col2 equals 3. I assume this would cause an index > scan because the executor doesn't know about the most common value, > right? Is it worth trying to improve that? Oh, I see: you are assuming that a nestloop join is being done, and wondering if it's worthwhile to switch dynamically between seqscan and indexscan for each scan of the inner relation, depending on exactly what value is being supplied from the outer relation for that scan. Hmm. Not sure if it's worth the trouble or not. Nestloop is usually a last-resort join strategy anyway, and is unlikely to be picked when the tables are large enough to make performance be a big issue. regards, tom lane From tgl@sss.pgh.pa.us Mon Oct 16 01:48:27 2000 Received: from sss.pgh.pa.us (sss.pgh.pa.us [209.114.132.154]) by candle.pha.pa.us (8.9.0/8.9.0) with ESMTP id AAA01602 for ; Mon, 16 Oct 2000 00:48:26 -0400 (EDT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.11.1/8.11.1) with ESMTP id e9G4mu521809; Mon, 16 Oct 2000 00:48:56 -0400 (EDT) To: Bruce Momjian cc: Jules Bean , Alfred Perlstein , pgsql-hackers@postgresql.org Subject: Re: [HACKERS] Performance on inserts In-reply-to: <200010160441.AAA01374@candle.pha.pa.us> References: <200010160441.AAA01374@candle.pha.pa.us> Comments: In-reply-to Bruce Momjian message dated "Mon, 16 Oct 2000 00:41:49 -0400" Date: Mon, 16 Oct 2000 00:48:56 -0400 Message-ID: <21806.971671736@sss.pgh.pa.us> From: Tom Lane Status: OR Bruce Momjian writes: >> So an inner indexscan for tab1 is definitely a possible plan. > Yes, that was my point, that a nested loop could easily be involved if > the joined table has a restriction. Is there a TODO item here? More like a "to investigate" --- I'm not sold on the idea that a dynamic switch in plan types would be a win. Maybe it would be, but... One thing to think about is that it'd be critically dependent on having accurate statistics. Currently, the planner only places bets on the average behavior over a whole join. If you make a separate bet on each scan, then you open up the risk of betting wrong every time, should your stats be out-of-date or otherwise misleading. regards, tom lane From pgsql-hackers-owner+M29943@postgresql.org Thu Oct 3 18:18:27 2002 Return-path: Received: from postgresql.org (postgresql.org [64.49.215.8]) by candle.pha.pa.us (8.11.6/8.10.1) with ESMTP id g93MIOU23771 for ; Thu, 3 Oct 2002 18:18:25 -0400 (EDT) Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id B9F51476570; Thu, 3 Oct 2002 18:18:21 -0400 (EDT) Received: from postgresql.org (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with SMTP id E083B4761B0; Thu, 3 Oct 2002 18:18:19 -0400 (EDT) Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 13ADC476063 for ; Thu, 3 Oct 2002 18:18:17 -0400 (EDT) Received: from acorn.he.net (acorn.he.net [64.71.137.130]) by postgresql.org (Postfix) with ESMTP id 3AEC8475FFF for ; Thu, 3 Oct 2002 18:18:16 -0400 (EDT) Received: from CurtisVaio ([63.164.0.47] (may be forged)) by acorn.he.net (8.8.6/8.8.2) with SMTP id PAA19215; Thu, 3 Oct 2002 15:18:14 -0700 From: "Curtis Faith" To: "Tom Lane" cc: "Pgsql-Hackers" Subject: Re: [HACKERS] Advice: Where could I be of help? Date: Thu, 3 Oct 2002 18:17:55 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) In-Reply-To: <13379.1033675158@sss.pgh.pa.us> X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6700 Importance: Normal X-Virus-Scanned: by AMaViS new-20020517 Precedence: bulk Sender: pgsql-hackers-owner@postgresql.org X-Virus-Scanned: by AMaViS new-20020517 Status: OR tom lane wrote: > But more globally, I think that our worst problems these days have to do > with planner misestimations leading to bad plans. The planner is > usually *capable* of generating a good plan, but all too often it picks > the wrong one. We need work on improving the cost modeling equations > to be closer to reality. If that's at all close to your sphere of > interest then I think it should be #1 priority --- it's both localized, > which I think is important for a first project, and potentially a > considerable win. This seems like a very interesting problem. One of the ways that I thought would be interesting and would solve the problem of trying to figure out the right numbers is to have certain guesses for the actual values based on statistics gathered during vacuum and general running and then have the planner run the "best" plan. Then during execution if the planner turned out to be VERY wrong about certain assumptions the execution system could update the stats that led to those wrong assumptions. That way the system would seek the correct values automatically. We could also gather the stats that the system produces for certain actual databases and then use those to make smarter initial guesses. I've found that I can never predict costs. I always end up testing empirically and find myself surprised at the results. We should be able to make the executor smart enough to keep count of actual costs (or a statistical approximation) without introducing any significant overhead. tom lane also wrote: > There is no "cache flushing". We have a shared buffer cache management > algorithm that's straight LRU across all buffers. There's been some > interest in smarter cache-replacement code; I believe Neil Conway is > messing around with an LRU-2 implementation right now. If you've got > better ideas we're all ears. Hmmm, this is the area that I think could lead to huge performance gains. Consider a simple system with a table tbl_master that gets read by each process many times but with very infrequent inserts and that contains about 3,000 rows. The single but heavily used index for this table is contained in a btree with a depth of three with 20 - 8K pages in the first two levels of the btree. Another table tbl_detail with 10 indices that gets very frequent inserts. There are over 300,000 rows. Some queries result in index scans over the approximatley 5,000 8K pages in the index. There is a 40M shared cache for this system. Everytime a query which requires the index scan runs it will blow out the entire cache since the scan will load more blocks than the cache holds. Only blocks that are accessed while the scan is going will survive. LRU is bad, bad, bad! LRU-2 might be better but it seems like it still won't give enough priority to the most frequently used blocks. I don't see how it would do better for the above case. I once implemented a modified cache algorithm that was based on the clock algorithm for VM page caches. VM paging is similar to databases in that there is definite locality of reference and certain pages are MUCH more likely to be requested. The basic idea was to have a flag in each block that represented the access time in clock intervals. Imagine a clock hand sweeping across a clock, every access is like a tiny movement in the clock hand. Blocks that are not accessed during a sweep are candidates for removal. My modification was to use access counts to increase the durability of the more accessed blocks. Each time a block is accessed it's flag is shifted left (up to a maximum number of shifts - ShiftN ) and 1 is added to it. Every so many cache accesses (and synchronously when the cache is full) a pass is made over each block, right shifting the flags (a clock sweep). This can also be done one block at a time each access so the clock is directly linked to the cache access rate. Any blocks with 0 are placed into a doubly linked list of candidates for removal. New cache blocks are allocated from the list of candidates. Accesses of blocks in the candidate list just removes them from the list. An index root node page would likely be accessed frequently enough so that all it's bits would be set so it would take ShiftN clock sweeps. This algorithm increased the cache hit ratio from 40% to about 90% for the cases I tested when compared to a simple LRU mechanism. The paging ratio is greatly dependent on the ratio of the actual database size to the cache size. The bottom line that it is very important to keep blocks that are frequently accessed in the cache. The top levels of large btrees are accessed many hundreds (actually a power of the number of keys in each page) of times more frequently than the leaf pages. LRU can be the worst possible algorithm for something like an index or table scan of large tables since it flushes a large number of potentially frequently accessed blocks in favor of ones that are very unlikely to be retrieved again. tom lane also wrote: > This is an interesting area. Keep in mind though that Postgres is a > portable DB that tries to be agnostic about what kernel and filesystem > it's sitting on top of --- and in any case it does not run as root, so > has very limited ability to affect what the kernel/filesystem do. > I'm not sure how much can be done without losing those portability > advantages. The kinds of things I was thinking about should be very portable. I found that simply writing the cache in order of the file system offset results in very greatly improved performance since it lets the head seek in smaller increments and much more smoothly, especially with modern disks. Most of the time the file system will create files are large sequential bytes on the physical disks in order. It might be in a few chunks but those chunks will be sequential and fairly large. tom lane also wrote: > Well, not really all that isolated. The bottom-level index code doesn't > know whether you're doing INSERT or UPDATE, and would have no easy > access to the original tuple if it did know. The original theory about > this was that the planner could detect the situation where the index(es) > don't overlap the set of columns being changed by the UPDATE, which > would be nice since there'd be zero runtime overhead. Unfortunately > that breaks down if any BEFORE UPDATE triggers are fired that modify the > tuple being stored. So all in all it turns out to be a tad messy to fit > this in :-(. I am unconvinced that the impact would be huge anyway, > especially as of 7.3 which has a shortcut path for dead index entries. Well, this probably is not the right place to start then. - Curtis ---------------------------(end of broadcast)--------------------------- TIP 4: Don't 'kill -9' the postmaster From pgsql-hackers-owner+M29945@postgresql.org Thu Oct 3 18:47:34 2002 Return-path: Received: from postgresql.org (postgresql.org [64.49.215.8]) by candle.pha.pa.us (8.11.6/8.10.1) with ESMTP id g93MlWU26068 for ; Thu, 3 Oct 2002 18:47:32 -0400 (EDT) Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id F2AAE476306; Thu, 3 Oct 2002 18:47:27 -0400 (EDT) Received: from postgresql.org (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with SMTP id E7B5247604F; Thu, 3 Oct 2002 18:47:24 -0400 (EDT) Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 9ADCC4761A1 for ; Thu, 3 Oct 2002 18:47:18 -0400 (EDT) Received: from sss.pgh.pa.us (unknown [192.204.191.242]) by postgresql.org (Postfix) with ESMTP id DDB0B476187 for ; Thu, 3 Oct 2002 18:47:17 -0400 (EDT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.12.5/8.12.5) with ESMTP id g93MlIhR015091; Thu, 3 Oct 2002 18:47:18 -0400 (EDT) To: "Curtis Faith" cc: "Pgsql-Hackers" Subject: Re: [HACKERS] Advice: Where could I be of help? In-Reply-To: References: Comments: In-reply-to "Curtis Faith" message dated "Thu, 03 Oct 2002 18:17:55 -0400" Date: Thu, 03 Oct 2002 18:47:17 -0400 Message-ID: <15090.1033685237@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: by AMaViS new-20020517 Precedence: bulk Sender: pgsql-hackers-owner@postgresql.org X-Virus-Scanned: by AMaViS new-20020517 Status: OR "Curtis Faith" writes: > Then during execution if the planner turned out to be VERY wrong about > certain assumptions the execution system could update the stats that led to > those wrong assumptions. That way the system would seek the correct values > automatically. That has been suggested before, but I'm unsure how to make it work. There are a lot of parameters involved in any planning decision and it's not obvious which ones to tweak, or in which direction, if the plan turns out to be bad. But if you can come up with some ideas, go to it! > Everytime a query which requires the index scan runs it will blow out the > entire cache since the scan will load more blocks than the cache > holds. Right, that's the scenario that kills simple LRU ... > LRU-2 might be better but it seems like it still won't give enough priority > to the most frequently used blocks. Blocks touched more than once per query (like the upper-level index blocks) will survive under LRU-2. Blocks touched once per query won't. Seems to me that it should be a win. > My modification was to use access counts to increase the durability of the > more accessed blocks. You could do it that way too, but I'm unsure whether the extra complexity will buy anything. Ultimately, I think an LRU-anything algorithm is equivalent to a clock sweep for those pages that only get touched once per some-long-interval: the single-touch guys get recycled in order of last use, which seems just like a clock sweep around the cache. The guys with some amount of preference get excluded from the once-around sweep. To determine whether LRU-2 is better or worse than some other preference algorithm requires a finer grain of analysis than this. I'm not a fan of "more complex must be better", so I'd want to see why it's better before buying into it ... > The kinds of things I was thinking about should be very portable. I found > that simply writing the cache in order of the file system offset results in > very greatly improved performance since it lets the head seek in smaller > increments and much more smoothly, especially with modern disks. Shouldn't the OS be responsible for scheduling those writes appropriately? Ye good olde elevator algorithm ought to handle this; and it's at least one layer closer to the actual disk layout than we are, thus more likely to issue the writes in a good order. It's worth experimenting with, perhaps, but I'm pretty dubious about it. BTW, one other thing that Vadim kept saying we should do is alter the cache management strategy to retain dirty blocks in memory (ie, give some amount of preference to as-yet-unwritten dirty pages compared to clean pages). There is no reliability cost here since the WAL will let us reconstruct any dirty pages if we crash before they get written; and the periodic checkpoints will ensure that we eventually write a dirty block and thus it will become available for recycling. This seems like a promising line of thought that's orthogonal to the basic LRU-vs-whatever issue. Nobody's got round to looking at it yet though. I've got no idea how much preference should be given to a dirty block --- not infinite, probably, but some. regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html From pgsql-hackers-owner+M29974@postgresql.org Fri Oct 4 01:28:54 2002 Return-path: Received: from postgresql.org (postgresql.org [64.49.215.8]) by candle.pha.pa.us (8.11.6/8.10.1) with ESMTP id g945SpU13476 for ; Fri, 4 Oct 2002 01:28:52 -0400 (EDT) Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 63999476BB2; Fri, 4 Oct 2002 01:26:56 -0400 (EDT) Received: from postgresql.org (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with SMTP id BB7CA476B85; Fri, 4 Oct 2002 01:26:54 -0400 (EDT) Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 5FD7E476759 for ; Fri, 4 Oct 2002 01:26:52 -0400 (EDT) Received: from mclean.mail.mindspring.net (mclean.mail.mindspring.net [207.69.200.57]) by postgresql.org (Postfix) with ESMTP id 1F4A14766D8 for ; Fri, 4 Oct 2002 01:26:51 -0400 (EDT) Received: from 1cust163.tnt1.st-thomas.vi.da.uu.net ([200.58.4.163] helo=CurtisVaio) by mclean.mail.mindspring.net with smtp (Exim 3.33 #1) id 17xKzB-0000yK-00; Fri, 04 Oct 2002 01:26:49 -0400 From: "Curtis Faith" To: "Tom Lane" cc: "Pgsql-Hackers" Subject: Re: [HACKERS] Advice: Where could I be of help? Date: Fri, 4 Oct 2002 01:26:36 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) In-Reply-To: <15090.1033685237@sss.pgh.pa.us> X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6700 Importance: Normal X-Virus-Scanned: by AMaViS new-20020517 Precedence: bulk Sender: pgsql-hackers-owner@postgresql.org X-Virus-Scanned: by AMaViS new-20020517 Status: OR I wrote: > > My modification was to use access counts to increase the > durability of the > > more accessed blocks. > tom lane replies: > You could do it that way too, but I'm unsure whether the extra > complexity will buy anything. Ultimately, I think an LRU-anything > algorithm is equivalent to a clock sweep for those pages that only get > touched once per some-long-interval: the single-touch guys get recycled > in order of last use, which seems just like a clock sweep around the > cache. The guys with some amount of preference get excluded from the > once-around sweep. To determine whether LRU-2 is better or worse than > some other preference algorithm requires a finer grain of analysis than > this. I'm not a fan of "more complex must be better", so I'd want to see > why it's better before buying into it ... I'm definitely not a fan of "more complex must be better either". In fact, its surprising how often the real performance problems are easy to fix and simple while many person years are spent solving the issue everyone "knows" must be causing the performance problems only to find little gain. The key here is empirical testing. If the cache hit ratio for LRU-2 is much better then there may be no need here. OTOH, it took less than less than 30 lines or so of code to do what I described, so I don't consider it too, too "more complex" :=} We should run a test which includes running indexes (or is indices the PostgreSQL convention?) that are three or more times the size of the cache to see how well LRU-2 works. Is there any cache performance reporting built into pgsql? tom lane wrote: > Shouldn't the OS be responsible for scheduling those writes > appropriately? Ye good olde elevator algorithm ought to handle this; > and it's at least one layer closer to the actual disk layout than we > are, thus more likely to issue the writes in a good order. It's worth > experimenting with, perhaps, but I'm pretty dubious about it. I wasn't proposing anything other than changing the order of the writes, not actually ensuring that they get written that way at the level you describe above. This will help a lot on brain-dead file systems that can't do this ordering and probably also in cases where the number of blocks in the cache is very large. On a related note, while looking at the code, it seems to me that we are writing out the buffer cache synchronously, so there won't be any possibility of the file system reordering anyway. This appears to be a huge performance problem. I've read claims in the archives that that the buffers are written asynchronously but my read of the code says otherwise. Can someone point out my error? I only see calls that ultimately call FileWrite or write(2) which will block without a O_NOBLOCK open. I thought one of the main reasons for having a WAL is so that you can write out the buffer's asynchronously. What am I missing? I wrote: > > Then during execution if the planner turned out to be VERY wrong about > > certain assumptions the execution system could update the stats > that led to > > those wrong assumptions. That way the system would seek the > correct values > > automatically. tom lane replied: > That has been suggested before, but I'm unsure how to make it work. > There are a lot of parameters involved in any planning decision and it's > not obvious which ones to tweak, or in which direction, if the plan > turns out to be bad. But if you can come up with some ideas, go to > it! I'll have to look at the current planner before I can suggest anything concrete. - Curtis ---------------------------(end of broadcast)--------------------------- TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org