Ankit Jain

It was made default in [LUCENE-9398](https://issues.apache.org/jira/browse/LUCENE-9398) from 8.6. We need support in lucene that can be exposed as setting in Opensearch for moving it on/off heap as per the use...

> Was swapping enabled? No, swapping is disabled due to performance degradation concerns > We're ulimits set to unlimited? Yes they are: ``` % cat '/etc/security/limits.d/10-memlock.conf' * hard memlock unlimited...

@Pranshu-S - @peteralfonsi has done basic implementation in #13237 for avoiding this issue with cache stats. We can leverage something similar or come up with better way to achieve this.

@peternied - Thank you for reviewing the proposal and providing detailed feedback. Please find response to some of the questions below: > Clusters evolve over time by scaling horizontally (additional...

Thanks @deshsidd for writing this proposal. Most of it looks good, and storing hashes in memory is really nifty. Should prevent unnecessary memory consumption. Some considerations below: > Order of...

> @bowenlan-amzn could you also post the benchmark results of 2.11, where we didn't have any of these optimizations. +1. Will be good to look at the benchmark numbers before...

Also, I notice that the codecov/patch validation is failing due to low coverage. Can we cover few more use cases to increase the coverage to 75/80%?

> For each workload, old optimized results firt, then the new opimitzed results with multi range traversal, last one if exists would be the optimization disabled results which falls back...

Assigning to @kaushalmahi12, due to his prior context with query sandboxing and search backpressure

@peternied - Thank you for proposing this idea. While some of the aspects around access control / security make sense to me, I am unable to reason other benefits of...