Ben Ye

Consider query function when auto downsample

7

* [ ] I added CHANGELOG entry for this change. * [ ] Change is not relevant to the end user. ## Changes When auto downsample is enabled, trying to...

Remove obsolete state migration step in AM

1

**What this PR does**: `migrateStateFilesToPerTenantDirectories` has a TODO to remove this in Cortex 1.11. This pr removes the migration step. **Which issue(s) this PR fixes**: Fixes # **Checklist** - [...

Support time partitioned store gateway

5

**Is your feature request related to a problem? Please describe.** Thanos supports something called `time based partitioning` https://thanos.io/tip/components/store.md/#time-based-partitioning on Store Gateway. This is a good way to do hot/cold data...

Introduce index cache tracing

1

**What this PR does**: Instrument index cache with traces **Which issue(s) this PR fixes**: Fixes # **Checklist** - [ ] Tests updated - [ ] Documentation added - [x] `CHANGELOG.md`...

Replace sort.Slice with slices.SortFunc

2

**What this PR does**: I see `sort.Slice` under `sortLabelsIfNeeded` taking 3% CPU time of distributor. Similar idea of https://github.com/prometheus/prometheus/pull/12539 from upstream prometheus as the new function should have a better...

**Describe the bug** GH action run: https://github.com/cortexproject/cortex/actions/runs/6922198043/job/18828645369 ``` --- FAIL: TestDistributor_Push_RelabelDropWillExportMetricOfDroppedSamples (0.02s) distributor_test.go:3367: Error Trace: /__w/cortex/cortex/pkg/distributor/distributor_test.go:3367 Error: Received unexpected error: Diff: --- metric output does not match expectation; want +++...

QueryFrontend error cannot be parsed correctly in Grafana

3

**Describe the bug** When query frontend throws error, Grafana will show error message like below. ``` readObjectStart: expect { or n, but found z, error found in #1 byte of...

**What this PR does**: This pr separates the query limits for SG, mainly series and chunks limit. Previously we use the same limits from querier for store gateways. That works...

Make per tenant queue size reloadable during runtime

3

**Is your feature request related to a problem? Please describe.** We make max queue size a per tenant value in https://github.com/cortexproject/cortex/pull/4991. However, although they are per-tenant now, they are not...

Automated performance benchmarks for Cortex

10

**Is your feature request related to a problem? Please describe.** Right now, we don't have a way to identify and prevent performance regressions in Cortex. The reason is that we...