[
https://issues.apache.org/jira/browse/CASSANDRA-15907?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17148207#comment-17148207
]
Caleb Rackliffe commented on CASSANDRA-15907:
---------------------------------------------
I've posted [some WIP|https://github.com/apache/cassandra/pull/659] on the raft
of minor changes suggested above.
I've also had some offline conversation w/ [~cscotta] and [~adelapena] around
how we implement guardrails. One way forward might be to have two thresholds w/
different levels of enforcement (again partly taking inspiration from what we
do w/ tombstones). The first threshold would drop a warning in the logs to make
it plain that filtering protection is starting to encounter a significant
number of potentially stale replica results ("silent" replicas) and is
therefore keeping many more cached partial results on the heap than we would in
the optimal case. The second threshold would be where we start to fail queries.
If we base these on the number of materialized rows, there are at least two
ways to quantify things. The first is just to use an absolute threshold for
rows per query. The second is to determine an "expansion factor" or how many
rows we can cache as a factor of the provided query limit or page size
(whichever is lower). The former is simpler and probably more intuitive (and
would do its job even if a query used an enormous LIMIT), but the second
conceptually takes the user's intent around limits and page sizes into account.
> 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: 4.0-beta
>
>
> 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}}.
> *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: [email protected]
For additional commands, e-mail: [email protected]
