Getting Started on Joulie

Quickstart

Mon, 01 Jan 0001 00:00:00 +0000

Prerequisites

Kubernetes cluster with worker nodes
Node Feature Discovery (NFD) deployed
Access to push images to registry.cern.ch/mbunino/joulie
Optional for real enforcement: nodes exposing writable power interfaces
- RAPL power limit files, or
- cpufreq sysfs interfaces

1. Install from release (recommended)

Install directly from OCI chart release:

helm upgrade --install joulie oci://registry.cern.ch/mbunino/joulie/joulie \
 --version <version> \
 -n joulie-system \
 --create-namespace \
 -f values/joulie.yaml

2. Install from source (when developing)

If you changed source code, from repo root:

Agent Runtime Modes

Mon, 01 Jan 0001 00:00:00 +0000

The agent supports two runtime modes:

daemonset: real-hardware mode, one pod per real node.
pool: simulation mode, one pod hosts many logical per-node controllers.

Chart templates:

charts/joulie/templates/agent-daemonset.yaml
charts/joulie/templates/agent-statefulset.yaml

DaemonSet mode (real hardware)

Required runtime settings

securityContext.privileged: true
Host mount:
- host path /sys -> container path /host-sys
Env:
- NODE_NAME from spec.nodeName
- AGENT_MODE=daemonset (default)
- optional RECONCILE_INTERVAL (default 20s)
- optional SIMULATE_ONLY=true (skip host writes, log requested actions)
- optional METRICS_ADDR (default :8080)

Pool mode (KWOK / simulation)

Pool mode preserves per-node semantics but shards logical node controllers across replicas.

Pod Compatibility for Joulie

Mon, 01 Jan 0001 00:00:00 +0000

Joulie uses Kubernetes scheduling constraints as the single source of truth for workload placement intent.

Power profile supply is exposed on node label:

joulie.io/power-profile=performance
joulie.io/power-profile=eco
joulie.io/power-profile=draining-performance (transition state)

Workload behavior:

performance workload: require joulie.io/power-profile=performance
eco workload: require joulie.io/power-profile=eco
unconstrained workload: no power-profile affinity, can run on either profile

Base Pod Spec

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17


apiVersion: apps/v1
kind: Deployment
metadata:
 name: my-workload
spec:
 replicas: 1
 selector:
 matchLabels:
 app: my-workload
 template:
 metadata:
 labels:
 app: my-workload
 spec:
 containers:
 - name: app
 image: ghcr.io/example/app:latest

Performance-only Pod (lines to add)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25


apiVersion: apps/v1
kind: Deployment
metadata:
 name: my-workload
spec:
 replicas: 1
 selector:
 matchLabels:
 app: my-workload
 template:
 metadata:
 labels:
 app: my-workload
 spec:
 affinity:
 nodeAffinity:
 requiredDuringSchedulingIgnoredDuringExecution:
 nodeSelectorTerms:
 - matchExpressions:
 - key: joulie.io/power-profile
 operator: In
 values: ["performance"]
 containers:
 - name: app
 image: ghcr.io/example/app:latest

Eco-only Pod (lines to add)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25


apiVersion: apps/v1
kind: Deployment
metadata:
 name: my-workload
spec:
 replicas: 1
 selector:
 matchLabels:
 app: my-workload
 template:
 metadata:
 labels:
 app: my-workload
 spec:
 affinity:
 nodeAffinity:
 requiredDuringSchedulingIgnoredDuringExecution:
 nodeSelectorTerms:
 - matchExpressions:
 - key: joulie.io/power-profile
 operator: In
 values: ["eco"]
 containers:
 - name: app
 image: ghcr.io/example/app:latest

Unconstrained Pod

Do not set power-profile affinity. Kubernetes can schedule it on either eco or performance nodes.