[ofa-general] Re: [Query] ib add path record cache

[Devesh Sharma]

On 5/24/07, Sean Hefty <mshefty at ichips.intel.com> wrote:
> > Yes It will, and hence reduce the initial SA traffic generated on a
> > big cluster...just imagin, the cluster is quite big and every node is
> > trying to build its cache initially. It will create large burst of SA
> > packets.
>
> In general I agree with the notion of enhancing the cache to allow it to
> load locally from a file.  But I'd really like to get a framework merged
> upstream first before trying to add in these sort of enhancements.
Ok, but, by that time we can keep the framework ready?
>
> Initially loading of caches on a large fabric can be limited to a single
> GetTable PR query per node, and by enabling the caches across the fabric
> at different times, the single burst to the SA can be avoided.
How this will be managed? This will add extra startup time in the
cluster, because cluster will be usable only after last cache has been
enabled. Am I right?
>
> > Incomplete connectivity will be till first PR is requested for that
> > destination, Because if its a cache miss, any how application is going
> > to initiate a ib_sa_get_path_rec() and resolved PR will be added in
> > cache for future reference.
>
> ib_sa_get_path_rec() only returns a single path.  If multiple paths
> exist, adding a single path to the cache may cause all applications to
How multi-pathing is handled in current cache_module?
> make use of that single path. Updating the cache on demand isn't as
> simple as it seems.
>
> - Sean
>