Hashed file performance issue

[dr.murthy]

Hi All,

Source Hash(SH)====>Xfm======>normalised Hash(NH)======>Target

i had a job its reading the data from universe hashed file(SH) and writing it to another
hashed file(NH), same hashed stage is normalising the data and load into final sequential
file.
But here i have around 3 billion records, job takes more that 2 hrs to complete.
Is there any best practices to improove the performance in this design.

Especially in normalised Hash(NH) , job is waiting till complete all the records write.

Appreciate your help on this issue.

[ray.wurlod]

Have you looked at tuning the hashed file, particularly reserving the disk space they need? Are the records very large?