Dmitry, I believe you're talking about this issue: https://issues.apache.org/jira/browse/IGNITE-1903
I just added a comment there with some thoughts about how we can improve this. -Val On Mon, Feb 20, 2017 at 12:37 AM, Alexey Goncharuk < alexey.goncha...@gmail.com> wrote: > Dmitriy, > > What do you mean by deploying cache metadata? The node filter itself must > be deserialized on each node because we need to evaluate it. As for the > CacheConfiguration, we do not need to deserialize the configuration on > every node, this will be a good change for Ignite 2.0. > > 2017-02-16 23:14 GMT+03:00 Dmitriy Setrakyan <dsetrak...@apache.org>: > > > Igniters, > > > > I believe that we are not propagating cache metadata correctly, but I > want > > to confirm it with the community first. > > > > Currently, it is possible in Ignite to split the cluster into logical > > groups, and deploy caches only on certain nodes using NodeFilters. For > > example, CacheA will reside only on Nodes 1, 2, and 3, and CacheB will > > reside only on nodes 4 and 5. However, even in this case, it looks like > the > > cache metadata for both caches will be deployed on all the nodes. > > > > It looks like the correct approach would be to honor the NodeFilter for a > > cache whenever deploying the cache metadata. Is there a reason for the > > metadata to be deployed on all the nodes? > > > > D. > > >
