database/sqlite
1
0
Fork 0
mirror of https://github.com/sqlite/sqlite.git synced 2025-11-18 10:21:03 +03:00
Code Activity

Reduce the N in the logN term for the sorting cost associated with computing

Browse Source 
DISTINCT by B-Tree by half, under the assumption that the DISTINCT will
eliminate about half the rows of output.  This is really a wild guess.  But
we do not have any better way of estimating what the row-count reduction due
to DISTINCT will actually be.

FossilOrigin-Name: 8787417ec1da8071d84c6ff0d7a90b5fd458ab6baba871327f36bc4e1bceca61
This commit is contained in:
drh
2020-08-24 23:44:27 +00:00
parent 599456f0ab
commit 58d6f633ba
3 changed files with 18 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
manifest
View File

@@ -1,5 +1,5 @@
C Correctly\shandle\sexpressions\slike\s"x\sIS\s(not)\strue/false"\swithin\sthe\srhs\sof\sIN()\sexpressions.\sFix\sfor\s[f3ff1472]. C Reduce\sthe\sN\sin\sthe\slogN\sterm\sfor\sthe\ssorting\scost\sassociated\swith\scomputing\nDISTINCT\sby\sB-Tree\sby\shalf,\sunder\sthe\sassumption\sthat\sthe\sDISTINCT\swill\neliminate\sabout\shalf\sthe\srows\sof\soutput.\s\sThis\sis\sreally\sa\swild\sguess.\s\sBut\nwe\sdo\snot\shave\sany\sbetter\sway\sof\sestimating\swhat\sthe\srow-count\sreduction\sdue\nto\sDISTINCT\swill\sactually\sbe.
D 2020-08-24T10:52:52.642 D 2020-08-24T23:44:27.235
F .fossil-settings/empty-dirs dbb81e8fc0401ac46a1491ab34a7f2c7c0452f2f06b54ebb845d024ca8283ef1 F .fossil-settings/empty-dirs dbb81e8fc0401ac46a1491ab34a7f2c7c0452f2f06b54ebb845d024ca8283ef1
F .fossil-settings/ignore-glob 35175cdfcf539b2318cb04a9901442804be81cd677d8b889fcc9149c21f239ea F .fossil-settings/ignore-glob 35175cdfcf539b2318cb04a9901442804be81cd677d8b889fcc9149c21f239ea
F LICENSE.md df5091916dbb40e6e9686186587125e1b2ff51f022cc334e886c19a0e9982724 F LICENSE.md df5091916dbb40e6e9686186587125e1b2ff51f022cc334e886c19a0e9982724
@@ -622,7 +622,7 @@ F src/vxworks.h d2988f4e5a61a4dfe82c6524dd3d6e4f2ce3cdb9
F src/wal.c 69e770e96fd56cc21608992bf2c6f1f3dc5cf2572d0495c6a643b06c3a679f14 F src/wal.c 69e770e96fd56cc21608992bf2c6f1f3dc5cf2572d0495c6a643b06c3a679f14
F src/wal.h c3aa7825bfa2fe0d85bef2db94655f99870a285778baa36307c0a16da32b226a F src/wal.h c3aa7825bfa2fe0d85bef2db94655f99870a285778baa36307c0a16da32b226a
F src/walker.c 3df26a33dc4f54e8771600fb7fdebe1ece0896c2ad68c30ab40b017aa4395049 F src/walker.c 3df26a33dc4f54e8771600fb7fdebe1ece0896c2ad68c30ab40b017aa4395049
F src/where.c 396ba2c62defd56283ef54a4b221f698f4733fe8fbfca4aa8b1cc4f8c58d12ee F src/where.c 23f47e845e304a41d0b221bf67bd170014ae08b673076813fcd945dda1a3d4af
F src/whereInt.h eb8c2847fb464728533777efec1682b3c074224293b2da73513c61a609efbeab F src/whereInt.h eb8c2847fb464728533777efec1682b3c074224293b2da73513c61a609efbeab
F src/wherecode.c 110fa357bf453e0d30bf5ebb2cc86ea34a3631c39b857f30c228fd325cb53ae7 F src/wherecode.c 110fa357bf453e0d30bf5ebb2cc86ea34a3631c39b857f30c228fd325cb53ae7
F src/whereexpr.c 264d58971eaf8256eb5b0917bcd7fc7a1f1109fdda183a8382308a1b18a2dce7 F src/whereexpr.c 264d58971eaf8256eb5b0917bcd7fc7a1f1109fdda183a8382308a1b18a2dce7
@@ -1879,7 +1879,7 @@ F vsixtest/vsixtest.tcl 6a9a6ab600c25a91a7acc6293828957a386a8a93
F vsixtest/vsixtest.vcxproj.data 2ed517e100c66dc455b492e1a33350c1b20fbcdc F vsixtest/vsixtest.vcxproj.data 2ed517e100c66dc455b492e1a33350c1b20fbcdc
F vsixtest/vsixtest.vcxproj.filters 37e51ffedcdb064aad6ff33b6148725226cd608e F vsixtest/vsixtest.vcxproj.filters 37e51ffedcdb064aad6ff33b6148725226cd608e
F vsixtest/vsixtest_TemporaryKey.pfx e5b1b036facdb453873e7084e1cae9102ccc67a0 F vsixtest/vsixtest_TemporaryKey.pfx e5b1b036facdb453873e7084e1cae9102ccc67a0
P 5f58dd3a19605b6f49b4364fa29892502eff35f12a7693a8694100e1844711ea P 493a25949b9a6d0be82169b597133e491d8be4f4147b6eae135b06c1d810abd3
R e7a4532d833ea3a282e0c879ce028741 R ddd62187ac7223d5e6ce61819bf672f8
U dan U drh
Z b8cdd5c782cc19b36092e011278ca641 Z 3f2dc6629d15095497f0f8f6b9a935ba

2
manifest.uuid
View File

@@ -1 +1 @@
493a25949b9a6d0be82169b597133e491d8be4f4147b6eae135b06c1d810abd3 8787417ec1da8071d84c6ff0d7a90b5fd458ab6baba871327f36bc4e1bceca61

12
src/where.c
View File

@@ -4053,16 +4053,24 @@ static LogEst whereSortingCost(
** cost = (3.0 * N * log(N)) * (Y/X) ** cost = (3.0 * N * log(N)) * (Y/X)
** **
** The (Y/X) term is implemented using stack variable rScale ** The (Y/X) term is implemented using stack variable rScale
** below. */ ** below.
*/
LogEst rScale, rSortCost; LogEst rScale, rSortCost;
assert( nOrderBy>0 && 66==sqlite3LogEst(100) ); assert( nOrderBy>0 && 66==sqlite3LogEst(100) );
rScale = sqlite3LogEst((nOrderBy-nSorted)*100/nOrderBy) - 66; rScale = sqlite3LogEst((nOrderBy-nSorted)*100/nOrderBy) - 66;
rSortCost = nRow + rScale + 16; rSortCost = nRow + rScale + 16;
/* Multiple by log(M) where M is the number of output rows. /* Multiple by log(M) where M is the number of output rows.
** Use the LIMIT for M if it is smaller */ ** Use the LIMIT for M if it is smaller. Or if this sort is for
** a DISTINT operator, M will be the number of distinct output
** rows, so fudge it downwards a bit.
*/
if( (pWInfo->wctrlFlags & WHERE_USE_LIMIT)!=0 && pWInfo->iLimit<nRow ){ if( (pWInfo->wctrlFlags & WHERE_USE_LIMIT)!=0 && pWInfo->iLimit<nRow ){
nRow = pWInfo->iLimit; nRow = pWInfo->iLimit;
}else if( (pWInfo->wctrlFlags & WHERE_WANT_DISTINCT) ){
/* TUNING: In the sort for a DISTINCT operator, assume that the DISTINCT
** reduces the number of output rows by a factor of 2 */
if( nRow>10 ) nRow -= 10; assert( 10==sqlite3LogEst(2) );
} }
rSortCost += estLog(nRow); rSortCost += estLog(nRow);
return rSortCost; return rSortCost;