Could anyone please help me how to find the worst running queries lists in teradata?
Hope i will get the response soon....
or of course you could search these forums for articles like the following...http://developer.teradata.com/viewpoint/articles/killer-queries-track-them-find-them-fix-them
Some other calculations of interest include:
Product Join Indicator , results > 6 could be a performance issue, (TotalCPUTime *1000)/TotalIOCount = PJI. this may be distorted by large aggregations, numerous substring, case or index (not table index, but index("string","instring",startnum))functions.
Unnecesasry IO Indicator, TotalIOCount(TotalCPUTime*1000) , yes the inverse of the PJI function. results > 6 could indicate a missing index
ImpactCPU, MaxAmpCPUTime * (hashamp()+1) , the higher the result, the more impact the query has on the system. A primary metric for reviewing the ugly queries on a regular basis to see signs of improvement