Hi,

First of all, if you don't need ranges, you should use IgniteAtomicLong
instead of the sequence. But you should keep in mind that updating the
atomic long is actually the same as updating the same cache entry, so you
acquire an exclusive write lock each time you do this. That's why sequence
is much faster - it doesn't do any distributed operations within a reserved
range. If you need performance, you should revisit your approach and try
using sequences with larger ranges.

-Val



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/AtomicSequence-not-performing-well-with-2-nodes-cluster-set-up-tp9208p9237.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Reply via email to