[jira] [Commented] (CASSANDRA-11320) Improve backoff policy for cqlsh COPY FROM

["Adam Holmberg (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-11320?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15187745#comment-15187745 ] 

Adam Holmberg commented on CASSANDRA-11320:
-------------------------------------------

I agree with the general approach. I now see that the current implementation only stalls for a specific type of error (server-side timeouts), and does not halt new requests on availability exceptions, etc. I think it might also be wise to limit the total concurrent requests in-flight (per host?). It could be a high number -- just something to prevent runaways if the client is not the bottleneck.

> Improve backoff policy for cqlsh COPY FROM
> ------------------------------------------
>
>                 Key: CASSANDRA-11320
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-11320
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Tools
>            Reporter: Stefania
>            Assignee: Stefania
>             Fix For: 3.x
>
>
> Currently we have an exponential back-off policy in COPY FROM that kicks in when timeouts are received. However there are two limitations:
> * it does not cover new requests and therefore we may not back-off sufficiently to give time to an overloaded server to recover
> * the pause is performed in the receiving thread and therefore we may not process server messages quickly enough
> There is a static throttling mechanism in rows per second from feeder to worker processes (the INGESTRATE) but the feeder has no idea of the load of each worker process. However it's easy to keep track of how many chunks a worker process has yet to read by introducing a bounded semaphore.
> The idea is to move the back-off pauses to the worker processes main thread so as to include all messages, new and retries, not just the retries that timed out. The worker process will not read new chunks during the back-off pauses, and the feeder process can then look at the number of pending chunks before sending new chunks to a worker process.
> [~aholmber], [~aweisberg] what do you think?  



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)