[ZODB-Dev] performance of chunked data over ZEO

[Chris McDonough]

FWIW, whenever I've seen speed differences like this it has to do with
matching the application chunk size to the server/client chunk size.  I
don't know much about the ZEO protocol but I assume it uses an
asyncore-based mainloop on the server side and a dispatcher on the
client side that reads some sized chunk of data before doing something
with it.  Maybe you could play around with these chunksize values.

On Sun, 2004-11-28 at 14:56, Andreas Jung wrote:
> Inside an application that stored large amounts of data in the ZODB as 
> linked
> structure (very similar to the Pdata magic in OFS.Image) I discovered that
> the chunk size has a significant influcence on the performance.  Using a 
> chunk
> size of 16k (which is also the default in OFS.Image) I got the expected 
> transfer
> rate of about 2-3MB/second. Increasing the chunk size to 64K caused a 
> performance
> loss to about 100-150 KB/seconds. Any ideas why increasing the chunk size
> causes such a dramatic performance loss?
> 
> Andreas
> _______________________________________________
> For more information about ZODB, see the ZODB Wiki:
> http://www.zope.org/Wikis/ZODB/
> 
> ZODB-Dev mailing list  -  ZODB-Dev at zope.org
> http://mail.zope.org/mailman/listinfo/zodb-dev
>