Rick Steeves wrote: > On Wed, Jan 27, 2010 at 11:46 AM, Roger Heflin <[email protected]>wrote: > >> speeds, but is all how the mvpmc interfaced with the myth protocol. >>> Rick >>> >>> That tells me that the problem has nothing to do with CPU or disk access >> That would be network then. >> >> > I don't see how it's network. I'm moving effectively the same amount of data > across the network using either method (myth protocol or CIFS directly), > right? That it works with CIFS means that the network is handling the load > just fine, right? > > >> THe mvpmc does have issues if the block size is too large, it loses packets >> because the network chip is rather cheap. >> >> On nfs setting the blocksize above 4096 caused issues, I believe there is a >> similar parameter for the myth protocol setup, but I have never set it to >> don't know where it is. >> > > Do you mean blocksize of the file system? Because the size of the blocks on > the file system shouldn't affect network traffic really. (well I mean in the > grand scheme datacenters it does, but it shouldn't in this context.) > > The myth wiki talks about adjusting block size for XFS ( > http://www.mythtv.org/wiki/Optimizing_Performance), but again I'm not having > problems moving the data off the file system. > > The myth protocol wike (http://www.mythtv.org/wiki/Myth_Protocol) doesn't > talk about changing the payload/packet size. > > ? I don't see how using CIFS successfully doesn't remove the question of > the server's performance, and bring the entire issue back to the mvpmc code > that handles receipt of data from mythtv protocol. Given enough time (which > can be not much) it bombs out using the mythtv protocol. > > Rick >
No, that does not mean the network is good. The software can treat the network differently, bigger blocksize (packets going in/out at the same time) is generally faster if you have a network card that is up to it, it is much slower if you push the card too hard and it loses packets because the network cards internal buffers are too small for several packets. I believe the changes were on the mvpmc side. And it gets more critical the higher the bitrate is, and likely with the digital stuff the bitrate is probably higher than was being used with the hauppage cards, so is more of a problem, with the lower bitrates it can retransmit and still get that data there on time. ------------------------------------------------------------------------------ The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com _______________________________________________ Mvpmc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mvpmc-users mvpmc wiki: http://mvpmc.wikispaces.com/
