On 03/10/2011 07:00 AM, Ming Fu wrote: > It seems that without making the none cache-able reply bypass the store, > other > optimization is really not possible at this moment.
Yes, making a Store visit optional is probably a required first step for the low-level performance optimizations you mentioned, even if you think that non-caching traffic needs its own strand. >> Moreover, we have already done pretty much the same thing for requests: >> The server-side code can receive requests from multiple sources >> (client-side, ICAP, eCAP), without really knowing where the request is >> coming from. I believe the same should be done for handling traffic in >> the opposite direction. > >> Many pieces of the required interface are already implemented and can be >> reused. If you want to work on this, let's discuss specifics! > > Let me go through the code to have a feel of the scope. FWIW, the interfaces of interest are: StoreClient/store_client, BodyPipe, Server(ServerStateData), and Adaptation::Initiator. There is also ClientStreamAPI that you will have to deal with. Making that low-level byte/pointer manipulator to work with the above high-level APIs may be one of the trickiest parts of the project. Cheers, Alex.
