--Hi all,
We are quite happy to find out the Schema Registry(SR) feature provided by Confluent, and now we are trying to test the performance. But just want to figure out if it have a caching system on the client side.
My concern is, if SR and REST Proxy are installed on different machines(and I believe they should), my understanding is every RESTful call to the REST Proxy will call SR service once to get the schema info. So with large amount of REST call, the connection between Rest Proxy and SR could be a bottleneck. My guess is PROXY might have a local schema cache so when there is a REST Proxy call, so it does not need to call the SR service in another machine.
Base on my reading of the architect(http://docs.confluent.io/2.0.0/schema-registry/docs/design.html#kafka-backend), I didn’t see any caching system except the master/slave mode, but as I check in the source code, it seems it have a server-client mode.
io.confluent.kafka.schemaregistry.client.rest
io.confluent.kafka.schemaregistry.rest
So I guess there will be a SR client installed(on the REST Proxy side?) to act as the SR cache? In that case both the server side and client side have a restful service for SR? I also tried to find info regarding this from the installation part but with no luck.
Could you help with my questions? Appreciate if you guys could provide some additional info regarding the mechanism related to SR performance.
Thanks,
Jimmy
You received this message because you are subscribed to the Google Groups "Confluent Platform" group.
To unsubscribe from this group and stop receiving emails from it, send an email to confluent-platf...@googlegroups.com.
To post to this group, send email to confluent...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/confluent-platform/518b46f3-f888-41ca-bbbe-5f140c18d15e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Hi Ewen,Thanks much for the help! Now I think I understand how it works.So following code is the cache for Avro serializer/deserializer, which sits on the REST Proxy server side, and it should have a mapping of schema and schema_id(generated by SchemaRepository), may I know if I get it right?Thanks again!Regards,Zhiming(Jimmy) ZhengData Engineer, DS&E
--
You received this message because you are subscribed to a topic in the Google Groups "Confluent Platform" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/confluent-platform/qh5PGEOIPkQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to confluent-platf...@googlegroups.com.
To post to this group, send email to confluent...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/confluent-platform/CAE1jLMNj6sDvPHMkxyy5wB6c-ZibbR96W83XKoPzL-pjpOYORw%40mail.gmail.com.
--
You received this message because you are subscribed to the Google Groups "Confluent Platform" group.
To unsubscribe from this group and stop receiving emails from it, send an email to confluent-platf...@googlegroups.com.
To post to this group, send email to confluent...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/confluent-platform/CA%2BDCjejHKV4M-eWd%2BWgbuD7tFvADHvYxvc8noLYiHL2gjZbbxQ%40mail.gmail.com.