database/mariadb-columnstore-engine
1
0
Fork 0
mirror of https://github.com/mariadb-corporation/mariadb-columnstore-engine.git synced 2025-06-12 05:01:56 +03:00
Code Activity

This patch introduces support for scanning/filtering vectorized execution for numeric-based

Browse Source 
data types TEXT, CHAR, VARCHAR, FLOAT and DOUBLE are not yet supported by vectorized path
This patch introduces an example for Google benchmarking suite to measure a perf diff
b/w legacy scan/filtering code and the templated version
This commit is contained in:
Roman Nozdrin
2021-09-08 17:59:20 +00:00
committed by Roman Nozdrin
parent cac23b0afc
commit af36f9940f
22 changed files with 2720 additions and 143 deletions
Download Patch File Download Diff File Expand all files Collapse all files

80
writeengine/wrapper/writeengine.cpp
View File

@ -26,6 +26,7 @@
#include <unistd.h>
#include <boost/scoped_array.hpp>
#include <boost/scoped_ptr.hpp>
#include <unordered_map>
using namespace std;
#include "joblisttypes.h"
@ -70,6 +71,7 @@ namespace WriteEngine
{
StopWatch timer;
using OidToIdxMap = std::unordered_map<OID, ColStructList::size_type>;
/**@brief WriteEngineWrapper Constructor
*/
@ -1323,7 +1325,7 @@ int WriteEngineWrapper::insertColumnRecs(const TxnID& txnid,
// debug information for testing
if (isDebug(DEBUG_2))
{
printInputValue(colStructList, colValueList, ridList);
printInputValue(colStructList, colValueList, ridList, dctnryStructList, dictStrList);
}
// end
@ -2869,7 +2871,7 @@ int WriteEngineWrapper::insertColumnRec_SYS(const TxnID& txnid,
// debug information for testing
if (isDebug(DEBUG_2))
{
printInputValue(colStructList, colValueList, ridList);
printInputValue(colStructList, colValueList, ridList, dctnryStructList, dictStrList);
}
// end
@ -3531,7 +3533,7 @@ int WriteEngineWrapper::insertColumnRec_Single(const TxnID& txnid,
// debug information for testing
if (isDebug(DEBUG_2))
{
printInputValue(colStructList, colValueList, ridList);
printInputValue(colStructList, colValueList, ridList, dctnryStructList, dictStrList);
}
//Convert data type and column width to write engine specific
@ -4188,35 +4190,41 @@ int WriteEngineWrapper::insertColumnRec_Single(const TxnID& txnid,
***********************************************************/
void WriteEngineWrapper::printInputValue(const ColStructList& colStructList,
const ColValueList& colValueList,
const RIDList& ridList) const
const RIDList& ridList,
const DctnryStructList& dctnryStructList,
const DictStrList& dictStrList) const
{
ColTupleList curTupleList;
ColStruct curColStruct;
ColTuple curTuple;
string curStr;
ColStructList::size_type i;
ColTupleList::size_type j;
size_t j;
OidToIdxMap oidToIdxMap;
printf("\n=========================\n");
// printf("\nTable OID : %d \n", tableOid);
std::cerr << std::endl << "=========================" << std::endl;
printf("\nTotal RIDs: %zu\n", ridList.size());
std::cerr << "Total RIDs: " << ridList.size() << std::endl;
for (i = 0; i < ridList.size(); i++)
cout << "RID[" << i << "] : " << ridList[i] << "\n";
printf("\nTotal Columns: %zu\n", colStructList.size());
std::cerr << "RID[" << i << "] : " << ridList[i] << std::endl;
std::cerr << "Total Columns: " << colStructList.size() << std::endl;
for (i = 0; i < colStructList.size(); i++)
{
curColStruct = colStructList[i];
curTupleList = colValueList[i];
if (curColStruct.tokenFlag)
{
oidToIdxMap.insert({curColStruct.dataOid, i});
continue;
}
printf("\nColumn[%zu]", i);
printf("\nData file OID : %d \t", curColStruct.dataOid);
printf("\tWidth : %d \t Type: %d", curColStruct.colWidth, curColStruct.colDataType);
printf("\nTotal values : %zu \n", curTupleList.size());
std::cerr << "Column[" << i << "]";
std::cerr << "Data file OID : " << curColStruct.dataOid << "\t";
std::cerr << "Width : " << curColStruct.colWidth << "\t" << " Type: " << curColStruct.colDataType << std::endl;
std::cerr << "Total values : " << curTupleList.size() << std::endl;
for (j = 0; j < curTupleList.size(); j++)
{
@ -4226,19 +4234,22 @@ void WriteEngineWrapper::printInputValue(const ColStructList& colStructList,
{
if (curTuple.data.type() == typeid(int))
curStr = boost::lexical_cast<string>(boost::any_cast<int>(curTuple.data));
else if (curTuple.data.type() == typeid(unsigned int))
curStr = boost::lexical_cast<string>(boost::any_cast<unsigned int>(curTuple.data));
else if (curTuple.data.type() == typeid(float))
curStr = boost::lexical_cast<string>(boost::any_cast<float>(curTuple.data));
else if (curTuple.data.type() == typeid(long long))
curStr = boost::lexical_cast<string>(boost::any_cast<long long>(curTuple.data));
else if (curTuple.data.type() == typeid(unsigned long long))
curStr = boost::lexical_cast<string>(boost::any_cast<unsigned long long>(curTuple.data));
else if (curTuple.data.type() == typeid(int128_t))
{
datatypes::TSInt128 val(boost::any_cast<int128_t>(curTuple.data));
curStr = val.toString();
}
curStr = datatypes::TSInt128(boost::any_cast<int128_t>(curTuple.data)).toString();
else if (curTuple.data.type() == typeid(double))
curStr = boost::lexical_cast<string>(boost::any_cast<double>(curTuple.data));
else if (curTuple.data.type() == typeid(short))
curStr = boost::lexical_cast<string>(boost::any_cast<short>(curTuple.data));
else if (curTuple.data.type() == typeid(unsigned short))
curStr = boost::lexical_cast<string>(boost::any_cast<unsigned short>(curTuple.data));
else if (curTuple.data.type() == typeid(char))
curStr = boost::lexical_cast<string>(boost::any_cast<char>(curTuple.data));
else
@ -4249,12 +4260,36 @@ void WriteEngineWrapper::printInputValue(const ColStructList& colStructList,
}
if (isDebug(DEBUG_3))
printf("Value[%zu] : %s\n", j, curStr.c_str());
std::cerr << "Value[" << j << "]: " << curStr.c_str() << std::endl;
}
}
for (i = 0; i < dctnryStructList.size(); ++i)
{
if (dctnryStructList[i].dctnryOid == 0)
continue;
std::cerr << "Dict[" << i << "]";
std::cerr << " file OID : " << dctnryStructList[i].dctnryOid << " Token file OID: " << dctnryStructList[i].columnOid << "\t";
std::cerr << "Width : " << dctnryStructList[i].colWidth << "\t" << " Type: " << dctnryStructList[i].fCompressionType << std::endl;
std::cerr << "Total values : " << dictStrList.size() << std::endl;
if (isDebug(DEBUG_3))
{
for (j = 0; j < dictStrList[i].size(); ++j)
{
// We presume there will be a value.
auto tokenOidIdx = oidToIdxMap[dctnryStructList[i].columnOid];
std::cerr << "string [" << dictStrList[i][j] << "]" << std::endl;
bool isToken = colStructList[tokenOidIdx].colType == WriteEngine::WR_TOKEN &&
colStructList[tokenOidIdx].tokenFlag;
if (isToken && !colValueList[tokenOidIdx][j].data.empty())
{
Token t = boost::any_cast<Token>(colValueList[tokenOidIdx][j].data);
std::cerr << "Token: block pos:[" << t.op << "] fbo: [" << t.fbo << "] bc: [" << t.bc << "]" << std::endl;
}
}
}
}
printf("\n=========================\n");
std::cerr << "=========================" << std::endl;
}
/***********************************************************
@ -6046,7 +6081,6 @@ int WriteEngineWrapper::tokenize(const TxnID& txnid, DctnryTuple& dctnryTuple, i
{
int cop = op(ct);
m_dctnry[cop]->setTransId(txnid);
//cout << "Tokenizing dctnryTuple.sigValue " << dctnryTuple.sigValue << endl;
return m_dctnry[cop]->updateDctnry(dctnryTuple.sigValue, dctnryTuple.sigSize, dctnryTuple.token);
}