[
https://issues.apache.org/jira/browse/CASSANDRA-7937?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14133862#comment-14133862
]
Benedict commented on CASSANDRA-7937:
-------------------------------------
This should certainly be dealt with by the cluster. We cannot rely on well
behaved clients, and clients cannot easily calculate a safe data-rate cross
cluster, so any client change would at best help direct writes only, which with
RF>1 is not much help. Nor could it be as responsive.
My preferred solution to this is CASSANDRA-6812, which should keep the server
responding to writes within the timeout window even as it blocks for lengthy
flushes, but during these windows writes would be acked much more slowly, at a
steady drip. This solution won't make it into 2.0 or 2.1, and possibly not even
3.0, though.
> Apply backpressure gently when overloaded with writes
> -----------------------------------------------------
>
> Key: CASSANDRA-7937
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7937
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Environment: Cassandra 2.0
> Reporter: Piotr Kołaczkowski
> Labels: performance
>
> When writing huge amounts of data into C* cluster from analytic tools like
> Hadoop or Apache Spark, we can see that often C* can't keep up with the load.
> This is because analytic tools typically write data "as fast as they can" in
> parallel, from many nodes and they are not artificially rate-limited, so C*
> is the bottleneck here. Also, increasing the number of nodes doesn't really
> help, because in a collocated setup this also increases number of
> Hadoop/Spark nodes (writers) and although possible write performance is
> higher, the problem still remains.
> We observe the following behavior:
> 1. data is ingested at an extreme fast pace into memtables and flush queue
> fills up
> 2. the available memory limit for memtables is reached and writes are no
> longer accepted
> 3. the application gets hit by "write timeout", and retries repeatedly, in
> vain
> 4. after several failed attempts to write, the job gets aborted
> Desired behaviour:
> 1. data is ingested at an extreme fast pace into memtables and flush queue
> fills up
> 2. after exceeding some memtable "fill threshold", C* applies rate limiting
> to writes - the more the buffers are filled-up, the less writes/s are
> accepted, however writes still occur within the write timeout.
> 3. thanks to slowed down data ingestion, now flush can happen before all the
> memory gets used
> Of course the details how rate limiting could be done are up for a discussion.
> It may be also worth considering putting such logic into the driver, not C*
> core, but then C* needs to expose at least the following information to the
> driver, so we could calculate the desired maximum data rate:
> 1. current amount of memory available for writes before they would completely
> block
> 2. total amount of data queued to be flushed and flush progress (amount of
> data to flush remaining for the memtable currently being flushed)
> 3. average flush write speed
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
