perf(storagenode): check log level

[ijsong]

What this PR does

This PR slightly improves the storage node's performance by checking the log level, making the
logger not make slices in advance.

[ijsong]

Current dependencies on/for this PR:

• main
  • PR build(deps): bump go.etcd.io/etcd from v0.0.0-20211015134708-72d3e382e73c to v0.5.0-alpha.5.0.20230414071934-94593e63d45c #444
    • PR perf(storagenode): check log level #411 👈
      • PR perf(storagenode): remove backup from append request #412
        • PR refactor(storagenode): simplify unary RPC logging #413
          • PR perf(storagenode): estimate the number of batchlets #414
            • PR chore(metarepos): do not use unnecessary labels #445
              • PR perf(metarepos): add mrpb.StorageNodeUncommitReport pool #446
                • PR feat(all): add flags for logger #447
                  • PR refactor(all): use atomic types #448
                  • PR feat(storagenode): change append rpc from unary to stream #449
                  • PR fix(storagenode): support partial success/failure for append #450
                  • PR feat(storagenode): add configurations for initial window size #451
                  • PR feat(storagenode): add pipelined Append RPC handler #457
                  • PR feat(client): add LogStreamAppender #459
                  • PR test(benchmark): support log stream appender #462

This comment was auto-generated by Graphite.

[hungryjang]

I don't know how much it helps performance, but it seems to be very poor in usability.

[ijsong]

I don't know how much it helps performance, but it seems to be very poor in usability.

Yeah, it is too micro-optimization to recognize it.

[ijsong]

@ijsong started a stack merge that includes this pull request via Graphite.

[ijsong]

@ijsong merged this pull request with Graphite.