diff options
| author | Kubernetes Prow Robot <k8s-ci-robot@users.noreply.github.com> | 2021-04-26 00:27:02 -0700 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2021-04-26 00:27:02 -0700 |
| commit | b729d27adef74bf819ade035b35666530c3140f7 (patch) | |
| tree | d5515de930cfab544c8bc9f2c4de808e13f0294f | |
| parent | e97e6dbf9ae673b97be797e0b1ef70da8d078eed (diff) | |
| parent | b1d6f5f9eafacbe4863d780f623089bd6a1646a4 (diff) | |
Merge pull request #5742 from warmchang/clusterloader2
🐳 Link to clusterloader2, and new docs repo of etcd.
| -rw-r--r-- | sig-scalability/configs-and-limits/provider-configs.md | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/sig-scalability/configs-and-limits/provider-configs.md b/sig-scalability/configs-and-limits/provider-configs.md index 518ce5f1..ec38e9a0 100644 --- a/sig-scalability/configs-and-limits/provider-configs.md +++ b/sig-scalability/configs-and-limits/provider-configs.md @@ -120,7 +120,7 @@ proposed</td> * Leader election results are non-deterministic on a typical cluster, and a config would be best served to be configured as worst-case. Not presently known whether there are performance impacts resulting from leader election resulting in either co-location or distribution of those components. -* Improving the cluster performance loading to match production deployment scenarios is critical on-going work, especially clusterloader: [https://git.k8s.io/perf-tests/clusterloader](https://git.k8s.io/perf-tests/clusterloader) +* Improving the cluster performance loading to match production deployment scenarios is critical on-going work, especially clusterloader2: [https://git.k8s.io/perf-tests/clusterloader2](https://git.k8s.io/perf-tests/clusterloader2) * Multi-zone / multi-az deployments are often used to manage large clusters, but for testing/scalability efforts the target is intentionally a single Availability Zone. This keeps greater consistency between environments that do and don’t support AZ-based deployments. Failures during scalability testing are outside the SIG charter. Protecting against network partitioning and improving total cluster availability (one of the key benefits to a multi-AZ strategy) are currently out scope for the Scalability SIG efforts. @@ -159,4 +159,4 @@ Motivated by many of the issues above, an alternative configuration is reasonabl CoreOS commentary on etcd sizing. -[https://github.com/coreos/etcd/blob/master/Documentation/op-guide/hardware.md](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/hardware.md) +[Hardware guidelines for administering etcd clusters](https://github.com/etcd-io/website/blob/master/content/docs/v3.4/op-guide/hardware.md) |