I wasn't actually referring to XMLRPC client-server specifically (I
don't see an immediate utility in "persistent iterators"). Rather, I'm
concerned that the current setup is contrary to the general pygr
ideology of delayed execution. This wasn't an issue before I started
working large datasets (next gen. sequencing). Here I see a dramatic
degradation of _responsiveness_ of NLMSA. I have to sit and wait for
about 2 minutes before any results start to emerge because they have
to be _all_ collected before I see any. This is why I'd like to have
an iterator.
This may also apply to XMLRPC. If you are serving over a slower
connection, the client waits for the query to complete and then waits
for data transfer over the network. With an iterator you can send
results as they emerge using non-blocking network writes, hence
generally reducing the amount of time the client waits before the
result is received due to pipelining.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"pygr-dev" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/pygr-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to