Re: [HACKERS] Analyzing foreign tables & memory problems

2012-05-15 Thread Noah Misch
On Mon, May 14, 2012 at 09:21:20AM +0200, Albe Laurenz wrote: > Noah Misch wrote: > > Just thinking out loud, we could provide an "extern Datum > AnalyzeWideValue;" > > and direct FDW authors to use that particular datum. It could look > like a > > toasted datum of external size WIDTH_THRESHOLD+1

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-05-14 Thread Albe Laurenz
Noah Misch wrote: > 1) Expose WIDTH_THRESHOLD in commands/vacuum.h and add documentation >so that the authors of foreign data wrappers are aware of the >problem and can avoid it on their side. >This would be quite simple. >> Seems reasonable. How would the FDW retu

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-05-13 Thread Noah Misch
On Wed, May 02, 2012 at 12:20:39PM +0200, Albe Laurenz wrote: > >>> 1) Expose WIDTH_THRESHOLD in commands/vacuum.h and add documentation > >>>so that the authors of foreign data wrappers are aware of the > >>>problem and can avoid it on their side. > >>>This would be quite simple. > >

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-05-02 Thread Albe Laurenz
I wrote: > Noah Misch wrote: >>> During ANALYZE, in analyze.c, functions compute_minimal_stats >>> and compute_scalar_stats, values whose length exceed >>> WIDTH_THRESHOLD (= 1024) are not used for calculating statistics >>> other than that they are counted as "too wide rows" and assumed >>> to be

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-04-30 Thread Albe Laurenz
Noah Misch wrote: >> During ANALYZE, in analyze.c, functions compute_minimal_stats >> and compute_scalar_stats, values whose length exceed >> WIDTH_THRESHOLD (= 1024) are not used for calculating statistics >> other than that they are counted as "too wide rows" and assumed >> to be all different. >

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-04-30 Thread Albe Laurenz
Tom Lane wrote: >>> I'm fairly skeptical that this is a real problem, and would prefer not >>> to complicate wrappers until we see some evidence from the field that >>> it's worth worrying about. >> If I have a table with 10 rows and default_statistics_target >> at 100, then a sample of 3

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-04-30 Thread Albe Laurenz
Simon Riggs wrote: >>> During ANALYZE, in analyze.c, functions compute_minimal_stats >>> and compute_scalar_stats, values whose length exceed >>> WIDTH_THRESHOLD (= 1024) are not used for calculating statistics >>> other than that they are counted as "too wide rows" and assumed >>> to be all differ

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-04-30 Thread Noah Misch
On Mon, Apr 30, 2012 at 12:27:45PM +0200, Albe Laurenz wrote: > During ANALYZE, in analyze.c, functions compute_minimal_stats > and compute_scalar_stats, values whose length exceed > WIDTH_THRESHOLD (= 1024) are not used for calculating statistics > other than that they are counted as "too wide row

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-04-30 Thread Simon Riggs
On Mon, Apr 30, 2012 at 3:24 PM, Tom Lane wrote: > "Albe Laurenz" writes: >> During ANALYZE, in analyze.c, functions compute_minimal_stats >> and compute_scalar_stats, values whose length exceed >> WIDTH_THRESHOLD (= 1024) are not used for calculating statistics >> other than that they are counte

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-04-30 Thread Tom Lane
"Albe Laurenz" writes: > Tom Lane wrote: >> I'm fairly skeptical that this is a real problem, and would prefer not >> to complicate wrappers until we see some evidence from the field that >> it's worth worrying about. > If I have a table with 10 rows and default_statistics_target > at 100, th

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-04-30 Thread Albe Laurenz
Tom Lane wrote: >> During ANALYZE, in analyze.c, functions compute_minimal_stats >> and compute_scalar_stats, values whose length exceed >> WIDTH_THRESHOLD (= 1024) are not used for calculating statistics >> other than that they are counted as "too wide rows" and assumed >> to be all different. >

Re: [HACKERS] Analyzing foreign tables & memory problems

2012-04-30 Thread Tom Lane
"Albe Laurenz" writes: > During ANALYZE, in analyze.c, functions compute_minimal_stats > and compute_scalar_stats, values whose length exceed > WIDTH_THRESHOLD (= 1024) are not used for calculating statistics > other than that they are counted as "too wide rows" and assumed > to be all different.

[HACKERS] Analyzing foreign tables & memory problems

2012-04-30 Thread Albe Laurenz
During ANALYZE, in analyze.c, functions compute_minimal_stats and compute_scalar_stats, values whose length exceed WIDTH_THRESHOLD (= 1024) are not used for calculating statistics other than that they are counted as "too wide rows" and assumed to be all different. This works fine with regular tabl