docs

Table of workloads

Workload/tooling	Short Description	Minimum Requirements
Tooling	Setup pbench instrumentation tools	Cluster-admin, Privileged Containers
Test	Test/Run your workload from ssh Container	Cluster-admin, Privileged Containers
Baseline	Baseline metrics capture	Tooling job*
Scale	Scales worker nodes	Cluster-admin
NodeVertical	Node Kubelet Density	Labeling Nodes
PodVertical	Max Pod Density	None
MasterVertical	Master Node Stress workload	None
HTTP	HTTP ingress TPS/Latency	None
Network	TCP/UDP Throughput/Latency	Labeling Nodes, See below
Deployments Per Namespace	Maximum Deployments	None
PVCscale	PVCScale test	Working storageclass
Conformance	OCP/Kubernetes e2e tests	None
Namespaces per cluster	Maximum Namespaces	None
Services per namespace	Maximum services per namespace	None
FIO I/O test	FIO I/O test - stress storage backend	Privileged Containers, Working storage class

• Baseline job without a tooled cluster just idles a cluster. The goal is to capture resource consumption over a period of time to characterize resource requirements thus tooling is required. (For now)

Network

Test	Requirement
Pod to Pod	Labeling Nodes
Pod to Pod with HostNetwork	Labeling Nodes, Open firewall ports at Host
Pod to Service	Labeling Nodes

Workload Contributing Guidelines

See this page for workload contributing guidelines.

CI Pass/Fail

Each workload will implement a form of pass/fail criteria in order to flag if the tests have failed in CI.

Workload/tooling	Pass/Fail
Tooling	NA
Test	NA
Baseline	NA
Scale	Yes: Test Duration
NodeVertical	Yes: Exit Code, Test Duration
PodVertical	Yes: Exit Code, Test Duration
MasterVertical	Yes: Exit Code, Test Duration
HTTP	No
Network	No
Deployments Per Namespace	No
PVCscale	No
Conformance	No
Namespaces per cluster	Yes: Exit code, Test Duration
Services per namespace	Yes: Exit code, Test Duration
FIO I/O test	No