On Wed, Dec 17, 2008 at 4:12 PM, Salman Abdul Baset <[email protected]> wrote: > >>> It is not clear to me that space-wise, why per-hop reassembly of 2-3 1500 >>> bytes segment is better than storing a <10 byte message identifier and a >>> 4-byte IP address of the node from which the message was received. >>> >> >> The implication of the requirement that a peer MUST NOT interleave one >> message with another is that the amount of buffer required for >> reassembly is capped per connection. It's # of connections * MTU * >> max tolerated out-of-order arrival (which for TCP for example is 4). >> That's a deterministic amount of space that scales per connection, so >> it's pretty small. > > This approach creates head-of-line blocking. The storage space is needed for > reordering but more importantly, the space is needed to completely > reassemble the fragmented message at every hop as per above scheme. > > Per-hop reassembly has been brought up in the context of small number of > fragments (2-3) but what is this small really? Is 10 fragments small? Is 20 > fragments small? >
Like I said, consensus at MSP was to restrict the size of the messages in order to avoid needing to address the impact of HoL blocking and the other issues created by large messages. The max size would be determined by overlay configuration. If you believe the base protocol needs to address large message issues such as HoL blocking, please start a new thread to propose that. Bruce >> >>> Observe that return-path state or via-lists are required to maintain >>> symmetric routing. However, it is not clear to me why this should be a >>> MUST. >>> If sender->destination routing is done on object-ID and >>> destination->sender >>> routing is done on sender ID, then there is no need for via-lists or >>> return-path states. Such scheme utilizes existing connections between >>> peers. >> >> This has been discussed repeatedly on the mailing list and the >> discussion is in the appendix of the base draft. Please refer to >> those other discussions. > > > Indeed. My point was to highlight routing mode tradeoff in the context of > fragmentation discussion. > > -s > > _______________________________________________ P2PSIP mailing list [email protected] https://www.ietf.org/mailman/listinfo/p2psip
