Coherence Federation Cluster Using <max-bandwidth> in the Participant Configuration Without Units Is Invalid
(Doc ID 2522907.1)
Last updated on OCTOBER 20, 2023
Applies to:Oracle Coherence - Version 188.8.131.52.0 and later
Information in this document applies to any platform.
Coherence Federation Cluster1 and Cluster2 both have one machine each. Now when we are modifying an object with key k each 100 milliseconds, user is able to receive the updated value of key k after each 1000 milliseconds only and the intermediate values are skipped on the other cluster. For example if user updates the value for key k at each 100 millis to 1,2,3,4,5 ... 10,11,12 ... and so on. The values user receives on the other cluster are 1,1,1,....,10,10,10,...,20,20,20 and so on. Where all the get and set events were fired after 100 milli seconds. This means values for same keys are being updated after ~1 second. It seems a bit problematic for user as their SLA is 40 milli seconds. If updates on same key happen on every 10 milli seconds, federation stops, if update on same key happens 100 milli second local cluster gets updated remote cluster skips all the updates within ~1 second. The issue was with the <max-bandwidth> setting of "10". No. of units were specified resulting in the default units of "bits per second." This setting was copied from an example in the Coherence documentation.
To view full details, sign in with your My Oracle Support account.
Don't have a My Oracle Support account? Click to get started!
In this Document