On 10/26/2012 6:40 PM, Jeffrey Altman wrote: > The argument in favor of a TCP based solution is that RX cannot go fast > enough. Andrew's claim is that RX cannot use a window size greater than > 43.75K because of the 32 packet window limitation in 1.6. The fact is > that this limitation is not a protocol limitation but an implementation > limitation. Andrew points to Simon Wilkinson's past talks on RX as a > justification for this restriction. Of course, Simon's talks also > provide the road map for how to remove the bottlenecks and permit the > full window size of 256 packets to be used without performance > degradation. When combined with Jumbograms a window can support up to > 2MB of application data per call without any changes to the protocol.
A correction here. The maximum window size is 358KB with the existing limit of 1440 bytes per packet since the existing jumbo gram implementation combines multiple RX packets into a single transmission unit but does not increase the number of packets in the window. However, it is possible to implement a packet size negotiation in RX which would permit the maximum packet size to be increased to just under 64K. 64K packets combined with the current 256 packet window size can permit saturation of a 10Gbit pipe with a single RX call. Jeffrey Altman
signature.asc
Description: OpenPGP digital signature
