Skewness refers to the variation in the number of rows among the various AMPs. The higher the variation (some have high number of rows and some have very much less number of rows) then it is said to have high skewness. This affects the effective parallel execution in teradata. The way to get around is by assigning proper indexes to bring about even distribution of rows across the AMPs.
Skewfactor:
It tells the distribution of the rows.
-->If uniformly distributed(normal distribution) skew
factor is 0.
There is aformula to identify the skewness.
Please find the below.
SELECT TableName,SUM(CurrentPerm) AS
CurrentPerm,SUM(PeakPerm) AS PeakPerm,(100 -
(AVG(CurrentPerm)/MAX(CurrentPerm)*100)) AS SkewFactor
FROM Dbc.TableSize
WHERE DataBaseName = 'SALES'
GROUP BY 1 ORDER BY 1
Skewfactor:
It tells the distribution of the rows.
-->If uniformly distributed(normal distribution) skew
factor is 0.
There is aformula to identify the skewness.
Please find the below.
SELECT TableName,SUM(CurrentPerm) AS
CurrentPerm,SUM(PeakPerm) AS PeakPerm,(100 -
(AVG(CurrentPerm)/MAX(CurrentPerm)*100)) AS SkewFactor
FROM Dbc.TableSize
WHERE DataBaseName = 'SALES'
GROUP BY 1 ORDER BY 1
Post a Comment