xxq250
/
homework-jianmu
1
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

fix:[TD-32855] Improve percentile function when calculating on a small amount of data.

This commit is contained in:
Jing Sima 2024-11-11 16:39:43 +08:00
parent 1031a3487d
commit f7a2c857f4
3 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
source/libs/function/inc/tpercentile.h
View File

@ -26,7 +26,7 @@ extern "C" {
struct tMemBucket; struct tMemBucket;
int32_t tMemBucketCreate(int32_t nElemSize, int16_t dataType, double minval, double maxval, bool hasWindowOrGroup, int32_t tMemBucketCreate(int32_t nElemSize, int16_t dataType, double minval, double maxval, bool hasWindowOrGroup,
struct tMemBucket **pBucket); struct tMemBucket **pBucket, int32_t numOfElements);
void tMemBucketDestroy(struct tMemBucket **pBucket); void tMemBucketDestroy(struct tMemBucket **pBucket);

2
source/libs/function/src/builtinsimpl.c
View File

@ -1805,7 +1805,7 @@ int32_t percentileFunction(SqlFunctionCtx* pCtx) {
pResInfo->complete = true; pResInfo->complete = true;
return TSDB_CODE_SUCCESS; return TSDB_CODE_SUCCESS;
} else { } else {
code = tMemBucketCreate(pCol->info.bytes, type, pInfo->minval, pInfo->maxval, pCtx->hasWindowOrGroup, &pInfo->pMemBucket); code = tMemBucketCreate(pCol->info.bytes, type, pInfo->minval, pInfo->maxval, pCtx->hasWindowOrGroup, &pInfo->pMemBucket, pInfo->numOfElems);
if (TSDB_CODE_SUCCESS != code) { if (TSDB_CODE_SUCCESS != code) {
return code; return code;
} }

10
source/libs/function/src/tpercentile.c
View File

@ -269,18 +269,16 @@ static void resetSlotInfo(tMemBucket *pBucket) {
} }
int32_t tMemBucketCreate(int32_t nElemSize, int16_t dataType, double minval, double maxval, bool hasWindowOrGroup, int32_t tMemBucketCreate(int32_t nElemSize, int16_t dataType, double minval, double maxval, bool hasWindowOrGroup,
tMemBucket **pBucket) { tMemBucket **pBucket, int32_t numOfElements) {
*pBucket = (tMemBucket *)taosMemoryCalloc(1, sizeof(tMemBucket)); *pBucket = (tMemBucket *)taosMemoryCalloc(1, sizeof(tMemBucket));
if (*pBucket == NULL) { if (*pBucket == NULL) {
return terrno; return terrno;
} }
if (hasWindowOrGroup) { if (hasWindowOrGroup) {
// With window or group by, we need to shrink page size and reduce page num to save memory. // With window or group by, we need to shrink page size to save memory.
(*pBucket)->numOfSlots = DEFAULT_NUM_OF_SLOT / 8 ; // 128 bucket
(*pBucket)->bufPageSize = 4096; // 4k per page (*pBucket)->bufPageSize = 4096; // 4k per page
} else { } else {
(*pBucket)->numOfSlots = DEFAULT_NUM_OF_SLOT;
(*pBucket)->bufPageSize = 16384 * 4; // 16k per page (*pBucket)->bufPageSize = 16384 * 4; // 16k per page
} }
@ -302,6 +300,8 @@ int32_t tMemBucketCreate(int32_t nElemSize, int16_t dataType, double minval, dou
} }
(*pBucket)->elemPerPage = ((*pBucket)->bufPageSize - sizeof(SFilePage)) / (*pBucket)->bytes; (*pBucket)->elemPerPage = ((*pBucket)->bufPageSize - sizeof(SFilePage)) / (*pBucket)->bytes;
(*pBucket)->numOfSlots = TMIN((int16_t)(numOfElements / ((*pBucket)->elemPerPage * 6)) + 1, DEFAULT_NUM_OF_SLOT);
(*pBucket)->comparFn = getKeyComparFunc((*pBucket)->type, TSDB_ORDER_ASC); (*pBucket)->comparFn = getKeyComparFunc((*pBucket)->type, TSDB_ORDER_ASC);
(*pBucket)->hashFunc = getHashFunc((*pBucket)->type); (*pBucket)->hashFunc = getHashFunc((*pBucket)->type);
@ -587,7 +587,7 @@ int32_t getPercentileImpl(tMemBucket *pMemBucket, int32_t count, double fraction
// try next round // try next round
tMemBucket *tmpBucket = NULL; tMemBucket *tmpBucket = NULL;
int32_t code = tMemBucketCreate(pMemBucket->bytes, pMemBucket->type, pSlot->range.dMinVal, pSlot->range.dMaxVal, int32_t code = tMemBucketCreate(pMemBucket->bytes, pMemBucket->type, pSlot->range.dMinVal, pSlot->range.dMaxVal,
false, &tmpBucket); false, &tmpBucket, pSlot->info.size);
if (TSDB_CODE_SUCCESS != code) { if (TSDB_CODE_SUCCESS != code) {
tMemBucketDestroy(&tmpBucket); tMemBucketDestroy(&tmpBucket);
return code; return code;