Multiplexing patterns seems like the right thing to do. Aside from not sharing rocksdb, having 300 separate operators also results in more threads, network connections, etc. That makes it all less efficient...
On Tue, May 2, 2017 at 6:06 PM, Aljoscha Krettek <aljos...@apache.org> wrote: > They can’t (with the current design of Flink) because each CEP pattern > get’s executed by a separate operator. > > We could think about doing multiplexing of several patterns inside one > operator. It’s what I hinted at earlier as a possible solution when I > mentioned that you could implement your own operator that keeps track of > the patterns and does the pattern matching. > > Best, > Aljoscha > > On 2. May 2017, at 18:00, Elias Levy <fearsome.lucid...@gmail.com> wrote: > > Any reason they can't share a single RocksDB state backend instance? > > > On Fri, Apr 28, 2017 at 8:44 AM, Aljoscha Krettek <aljos...@apache.org> > wrote: > >> The problem here is that this will try to open 300 RocksDB instances on >> each of the TMs (depending on how the parallelism is spread between the >> machines this could be more or less). As the exception says, this will open >> too many files because each RocksDB instance has a directory with several >> files in it. >> >> One possible solution would be to increase the limit on open files but I >> don’t think that opening 300 RocksDB instances on one machine is a good >> idea for any size of machine. I think with this many patterns you could >> start thinking about writing the pattern matching yourself and multiplexing >> the several patterns in one stateful function or operator. >> >> @Stefan, what do you think about having this many Rocks instances? >> > > > >
