DSXchange

Hi All,

We have a job which runs every hour, and populate data to hashed file.
This hashed file used as look up in other jobs.
The data volume for this hashed file is about 0.3 million, so our job clear this hashed file first and load complete data set (Clear File Before writing enable) - its take around 5 mins to do so.

So what here we thinking is:-

1) Just make a job to load to Hashed file for initial load
2) Make other job to load only incremental part of Data to hashed file(Hourly updated data only) - we need to insert only the data (Data volume - few hundreds).

But here our question is :-

-How hash algorithm is running?Is it run after every load to Hashed file
-Is it search for this incremental part of data in overflow file?(Clear File Before Writing - not enable)

Else is there any negative impacts to this append of Hashed file?


Thanks
-Patel

The hashing algorithm is performed while storing the data. This algorithm is utilized to decide where the bytes of data are stored.
Ken has a white paper on hashed files. It will give you a good understanding of what goes into the .OVER files.
As far as concerns about appending to the hashed file, no major worries there. As long as you perform your logic on what is supposed to go into the hashed file, you should be fine.
Also keep a check on how large the hashed file grows.

Thanks DSguru,

DSguru2B wrote: Ken has a white paper on hashed files.

Where can i found this??

patel_ds wrote:Thanks DSguru,

DSguru2B wrote: Ken has a white paper on hashed files.

Where can i found this??

This forum also has a search facility. You will find it ONLY if you search for it.

See, I just found it -
viewtopic.php?t=86252