Or Gerlitz wrote:
Why not having a failed lookup as the --only-- trigger to update the cache? so the cache contains only paths that were demanded by some consumer. What is implementation you were considering, is it an SA replica having all those paths whose sgid is the local node gid?
I view MPI as one of the primary reasons for having a cache. The cache is updated using an SA GET_TABLE request, which is more efficient than sending separate SA GET requests for each path record. Waiting for a failed lookup to create the initial cache would delay the startup time for apps wanting all-to-all connection establishment. In this case, we also get the side effect that the SA receives GET_TABLE requests from every node at roughly the same time.
Your assumption is correct. The implementation will contain copies of all path records whose SGID is a local node GID. (Currently it contains only a single path record per SGID/DGID, but that will be expanded.)
- Sean _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
