Question
Caching in Pega 7.3
What kind of caching mechanism hazelcast uses in PRPC 7.3 - is it distributed or replicated cache? Or is it configurable?
***Edited by Moderator Marissa to update categories***
Accepted Solution
Hi Sai,
The load balancing technique used in Pega facilitates the attempts to provide an even backlog of demand across multiple processors or production facilities.
So in a multinode environment the load is distributed in all nodes.
When you are using 5 nodes then, all the nodes will build the cache if they are present in the cluster or if they are up.
Regards,
Vikash
Hi Vikash,
Thanks for your reply. I noticed below information yesterday in PegaCLUSTER.log file -
[59fa5d9f42a5a746adca518bcb2a0ca2] [3.8] Added cache config: CacheConfig{name='INTERNAL_cachename', managerPrefix='/hz/', inMemoryFormat=BINARY, backupCount=1, hotRestart=HotRestartConfig{enabled=false, fsync=false}}
I guess the backupCount parameter is a proof that distributed feature is in use. I have some more questions -
Say there are 5 nodes in a cluster. We can configure it as follows -
a) One node will be the primary node and rest 4 will be the backup.
b) One node will be the primary node and from rest 4, we decided to use 2 nodes as backup and skip other 2 nodes from configuring any cache.
So, among point a and b, what is followed in PEGA? Or is it possible to configure both a) and b) .
Accepted Solution
Hi Sai,
The load balancing technique used in Pega facilitates the attempts to provide an even backlog of demand across multiple processors or production facilities.
So in a multinode environment the load is distributed in all nodes.
When you are using 5 nodes then, all the nodes will build the cache if they are present in the cluster or if they are up.
Regards,
Vikash
To just get an idea about different caches available in Pega go through below article
https://community.pega.com/sites/default/files/help_v719/procomhelpmain.htm
As your question is about how to maintain cache in multinode cluster go through below articles
https://community.pega.com/sites/default/files/help_v74/procomhelpmain.htm
https://docs-previous.pega.com/performance/load-balancing-multi-node-cluster
https://collaborate.pega.com/question/how-dose-multi-nodecluster-system-works
Embedded Hazelcast is the default clustering topology ,all the nodes will be joined in cluster to communicate with each other and it uses distributed caching .
To check how the caching works in hazelcast please go through below article
https://www.tothenew.com/blog/caching-what-why-and-how-with-hazelcast/
Hope it helps .
Thanks,
Arun