[
https://issues.apache.org/jira/browse/CASSANDRA-15907?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17159482#comment-17159482
]
Caleb Rackliffe edited comment on CASSANDRA-15907 at 7/17/20, 5:38 AM:
-----------------------------------------------------------------------
bq. the main advantage of this approach is that in the absence of conflicts
nothing needs to be cached
Even though partition-restricted queries without digest mismatches will skip
{{DataResolver}} entirely, this still helps in the case where we have a
mismatch, but start with a large number of conflict-free rows, correct? If so,
I think this is a benefit we should consider, given how likely we are to be
dealing with partition-restricted queries.
bq. Also we could also just let it unbounded to minimize the number of RFP
queries
The one thing that makes me a little uneasy about this is the extra logic we
need to enforce the "target cache size". I propose we avoid that, simply leave
the guardrails we've already got in place to avoid catastrophe (and excessive
RFP queries), and see if that means we can simplify what remains (like having
to clear the {{contents}} array list between batches). I'll try this and see
how it looks and pull it into the main 3.0 branch if it works.
Aside from that, I think the only remaining question would be verifying the
safety of [aggressively
clearing|https://github.com/apache/cassandra/pull/659/commits/30b8f4bebd95b3520b637d6d25d6bc16cb4d81a2]
{{responses}}.
was (Author: maedhroz):
bq. the main advantage of this approach is that in the absence of conflicts
nothing needs to be cached
Even though partition-restricted queries without digest mismatches will skip
{{DataResolver}} entirely, this still helps in the case where we have a
mismatch, but start with a large number of conflict-free rows, correct? If so,
I think this is a benefit we can't ignore, given how likely we are to be
dealing with partition-restricted queries.
bq. Also we could also just let it unbounded to minimize the number of RFP
queries
The one thing that makes me a little uneasy about this is the extra logic we
need to enforce the "target cache size". I propose we avoid that, simply leave
the guardrails we've already got in place to avoid catastrophe (and excessive
RFP queries), and see if that means we can simplify what remains (like having
to clear the {{contents}} array list between batches). I'll try this and see
how it looks and pull it into the main 3.0 branch if it works.
Aside from that, I think the only remaining question would be verifying the
safety of [aggressively
clearing|https://github.com/apache/cassandra/pull/659/commits/30b8f4bebd95b3520b637d6d25d6bc16cb4d81a2]
{{responses}}.
> Operational Improvements & Hardening for Replica Filtering Protection
> ---------------------------------------------------------------------
>
> Key: CASSANDRA-15907
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15907
> Project: Cassandra
> Issue Type: Improvement
> Components: Consistency/Coordination, Feature/2i Index
> Reporter: Caleb Rackliffe
> Assignee: Caleb Rackliffe
> Priority: Normal
> Labels: 2i, memory
> Fix For: 3.0.x, 3.11.x, 4.0-beta
>
> Time Spent: 4h 20m
> Remaining Estimate: 0h
>
> CASSANDRA-8272 uses additional space on the heap to ensure correctness for 2i
> and filtering queries at consistency levels above ONE/LOCAL_ONE. There are a
> few things we should follow up on, however, to make life a bit easier for
> operators and generally de-risk usage:
> (Note: Line numbers are based on {{trunk}} as of
> {{3cfe3c9f0dcf8ca8b25ad111800a21725bf152cb}}.)
> *Minor Optimizations*
> * {{ReplicaFilteringProtection:114}} - Given we size them up-front, we may be
> able to use simple arrays instead of lists for {{rowsToFetch}} and
> {{originalPartitions}}. Alternatively (or also), we may be able to null out
> references in these two collections more aggressively. (ex. Using
> {{ArrayList#set()}} instead of {{get()}} in {{queryProtectedPartitions()}},
> assuming we pass {{toFetch}} as an argument to {{querySourceOnKey()}}.)
> * {{ReplicaFilteringProtection:323}} - We may be able to use
> {{EncodingStats.merge()}} and remove the custom {{stats()}} method.
> * {{DataResolver:111 & 228}} - Cache an instance of
> {{UnaryOperator#identity()}} instead of creating one on the fly.
> * {{ReplicaFilteringProtection:217}} - We may be able to scatter/gather
> rather than serially querying every row that needs to be completed. This
> isn't a clear win perhaps, given it targets the latency of single queries and
> adds some complexity. (Certainly a decent candidate to kick even out of this
> issue.)
> *Documentation and Intelligibility*
> * There are a few places (CHANGES.txt, tracing output in
> {{ReplicaFilteringProtection}}, etc.) where we mention "replica-side
> filtering protection" (which makes it seem like the coordinator doesn't
> filter) rather than "replica filtering protection" (which sounds more like
> what we actually do, which is protect ourselves against incorrect replica
> filtering results). It's a minor fix, but would avoid confusion.
> * The method call chain in {{DataResolver}} might be a bit simpler if we put
> the {{repairedDataTracker}} in {{ResolveContext}}.
> *Testing*
> * I want to bite the bullet and get some basic tests for RFP (including any
> guardrails we might add here) onto the in-JVM dtest framework.
> *Guardrails*
> * As it stands, we don't have a way to enforce an upper bound on the memory
> usage of {{ReplicaFilteringProtection}} which caches row responses from the
> first round of requests. (Remember, these are later used to merged with the
> second round of results to complete the data for filtering.) Operators will
> likely need a way to protect themselves, i.e. simply fail queries if they hit
> a particular threshold rather than GC nodes into oblivion. (Having control
> over limits and page sizes doesn't quite get us there, because stale results
> _expand_ the number of incomplete results we must cache.) The fun question is
> how we do this, with the primary axes being scope (per-query, global, etc.)
> and granularity (per-partition, per-row, per-cell, actual heap usage, etc.).
> My starting disposition on the right trade-off between
> performance/complexity and accuracy is having something along the lines of
> cached rows per query. Prior art suggests this probably makes sense alongside
> things like {{tombstone_failure_threshold}} in {{cassandra.yaml}}.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
