[Draft] Parametrize 2 Js/Fs benchmarks

[mprimi]

DRAFT for early feedback

Parametrize 2 of the new JetStream FileStore benchmarks to collect more data by running multiple configurations similar-but-different from base-case.

Signed-off-by: Your Name marco@nats.io

Example results:

 go test ./server/ --run '^NOOP$' --bench 'Benchmark_FileStore(SelectMsgBlock|LoadNextMsgSameFilterAsStream)' --count 3
goos: darwin
goarch: arm64
pkg: github.com/nats-io/nats-server/v2/server
Benchmark_FileStoreSelectMsgBlock/blockSize:128,msgSize:100,msgCount:1000-8             94922935                12.50 ns/op     8001.87 MB/s
Benchmark_FileStoreSelectMsgBlock/blockSize:128,msgSize:100,msgCount:1000-8             95003413                12.47 ns/op     8019.95 MB/s
Benchmark_FileStoreSelectMsgBlock/blockSize:128,msgSize:100,msgCount:1000-8             96121112                12.46 ns/op     8028.46 MB/s
Benchmark_FileStoreSelectMsgBlock/blockSize:128,msgSize:1000,msgCount:1000-8            96347498                12.36 ns/op     80905.26 MB/s
Benchmark_FileStoreSelectMsgBlock/blockSize:128,msgSize:1000,msgCount:1000-8            94877598                12.41 ns/op     80564.23 MB/s
Benchmark_FileStoreSelectMsgBlock/blockSize:128,msgSize:1000,msgCount:1000-8            95427434                12.42 ns/op     80489.93 MB/s
Benchmark_FileStoreSelectMsgBlock/blockSize:128,msgSize:100,msgCount:10000-8            34678740                32.24 ns/op     3101.94 MB/s
Benchmark_FileStoreSelectMsgBlock/blockSize:128,msgSize:100,msgCount:10000-8            37309516                30.96 ns/op     3229.89 MB/s
Benchmark_FileStoreSelectMsgBlock/blockSize:128,msgSize:100,msgCount:10000-8            37618242                31.55 ns/op     3169.30 MB/s
Benchmark_FileStoreLoadNextMsgSameFilterAsStream/msgSize:10,msgCount:100000,subjects:1024-8              5034210               238.9 ns/op        41.86 MB/s
Benchmark_FileStoreLoadNextMsgSameFilterAsStream/msgSize:10,msgCount:100000,subjects:1024-8              4703763               235.0 ns/op        42.55 MB/s
Benchmark_FileStoreLoadNextMsgSameFilterAsStream/msgSize:10,msgCount:100000,subjects:1024-8              5018709               236.0 ns/op        42.37 MB/s
Benchmark_FileStoreLoadNextMsgSameFilterAsStream/msgSize:100,msgCount:1000,subjects:1024-8               5022440               236.1 ns/op       423.64 MB/s
Benchmark_FileStoreLoadNextMsgSameFilterAsStream/msgSize:100,msgCount:1000,subjects:1024-8               5030818               237.2 ns/op       421.66 MB/s
Benchmark_FileStoreLoadNextMsgSameFilterAsStream/msgSize:100,msgCount:1000,subjects:1024-8               5030712               235.8 ns/op       424.02 MB/s
Benchmark_FileStoreLoadNextMsgSameFilterAsStream/msgSize:10,msgCount:1000000,subjects:10240-8            5044974               235.8 ns/op        42.41 MB/s
Benchmark_FileStoreLoadNextMsgSameFilterAsStream/msgSize:10,msgCount:1000000,subjects:10240-8            5067390               236.9 ns/op        42.21 MB/s
Benchmark_FileStoreLoadNextMsgSameFilterAsStream/msgSize:10,msgCount:1000000,subjects:10240-8            5038150               236.7 ns/op        42.25 MB/s
PASS
ok      github.com/nats-io/nats-server/v2/server        50.270s

[derekcollison]

In general love the idea. For this benchmark, subjects and msgs diversity per block really don't matter. We need to do a proper maple tree for this, but for now lookup by sequence which this is testing the first step of that is a modified linear (if small) or binary search.

However, this parametrization will be awesome for subject addressing layer which is way more complicated and is focus for filestore v2.

[mprimi]

@derekcollison I'll proceed to parametrize the rest. Once that's done you can supply sensible parameter values that make the most sense to test

[mprimi]

n.b. this was always set to 1 in the original benchmark, now it rolls around, could also be random

[derekcollison]

If the value has meaning for the test or benchmark I try to call it out.

[mprimi]

Should the hardcoded 10 here also be randomized or vary?
Maybe (n % 100) if we want to keep it consistently small?

[derekcollison]

Not for this one.

[neilalexander]

Is this still relevant / for rebasing?