Skip to content

Commit f89e4e2

Browse files
committed
fix: tweaks to support running tests locally on apple silicon; probably arm generally
1 parent 183393d commit f89e4e2

23 files changed

+324
-3
lines changed

Diff for: tests/kind/profiles/non-gpu1/keep-it-simple.arm64

+25
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,25 @@
1+
{
2+
"name": "keep-it-simple.arm64",
3+
"creationTime": 1664149787016,
4+
"lastModifiedTime": 1676828268772,
5+
"lastUsedTime": 1680126266770,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Ray Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/qiskit\",\"Base image\":\"rayproject/ray:2.3.0-py38-aarch64\",\"Command line prefix\":\"python3 main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": "{\"Number of Workers\":1,\"CPUs per worker\":\"500m\",\"GPUs per worker\":0,\"Memory per worker\":\"1.5Gi\",\"Ephemeral Storage per worker\":\"5Gi\"}",
18+
"kubernetes/context": "kind-codeflare-test",
19+
"kubernetes/choose/ns": "default",
20+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
21+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
22+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
23+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Keep It Simple"
24+
}
25+
}

Diff for: tests/kind/profiles/non-gpu1/mcad-coscheduler.arm64

+33
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
{
2+
"name": "mcad-coscheduler",
3+
"creationTime": 1660657756574,
4+
"lastModifiedTime": 1664150418757,
5+
"lastUsedTime": 1677700501499,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Ray Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/qiskit\",\"Base image\":\"rayproject/ray:2.3.0-py38-aarch64\",\"Command line prefix\":\"python3 main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": {
18+
"Number of Workers": 1,
19+
"CPUs per worker": "200m",
20+
"GPUs per worker": 0,
21+
"Memory per worker": "1.25Gi",
22+
"Ephemeral Storage per worker": "5Gi"
23+
},
24+
"kubernetes/context": "kind-codeflare-test",
25+
"kubernetes/choose/ns": "default",
26+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
27+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
28+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
29+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Use the Multi-user Enhanced Kubernetes Scheduler",
30+
"kubernetes/mcad/choose/job-priority": "Default Priority",
31+
"kubernetes/mcad/choose/scheduler": "MCAD with the Advanced Coscheduler"
32+
}
33+
}

Diff for: tests/kind/profiles/non-gpu1/mcad-default.arm64

+27
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,27 @@
1+
{
2+
"name": "mcad-default.arm64",
3+
"creationTime": 1660657756574,
4+
"lastModifiedTime": 1660747919298,
5+
"lastUsedTime": 1680126433820,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Ray Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/qiskit\",\"Base image\":\"rayproject/ray:2.3.0-py38-aarch64\",\"Command line prefix\":\"python3 main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": "{\"Number of Workers\":1,\"CPUs per worker\":\"200m\",\"GPUs per worker\":0,\"Memory per worker\":\"1.25Gi\",\"Ephemeral Storage per worker\":\"5Gi\"}",
18+
"kubernetes/context": "kind-codeflare-test",
19+
"kubernetes/choose/ns": "default",
20+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
21+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
22+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
23+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Use the Multi-user Enhanced Kubernetes Scheduler",
24+
"kubernetes/mcad/choose/job-priority": "Default Priority",
25+
"kubernetes/mcad/choose/scheduler": "MCAD with the Default Kubernetes Scheduler"
26+
}
27+
}

Diff for: tests/kind/profiles/non-gpu1/mcad-preinstalled.arm64

+33
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
{
2+
"name": "mcad-preinstalled",
3+
"creationTime": 1660657756574,
4+
"lastModifiedTime": 1660747919298,
5+
"lastUsedTime": 1668210325310,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Ray Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/qiskit\",\"Base image\":\"rayproject/ray:2.3.0-py38-aarch64\",\"Command line prefix\":\"python3 main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": {
18+
"Number of Workers": 1,
19+
"CPUs per worker": "200m",
20+
"GPUs per worker": 0,
21+
"Memory per worker": "1.25Gi",
22+
"Ephemeral Storage per worker": "5Gi"
23+
},
24+
"kubernetes/context": "kind-codeflare-test",
25+
"kubernetes/choose/ns": "default",
26+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
27+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
28+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
29+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Use the Multi-user Enhanced Kubernetes Scheduler",
30+
"kubernetes/mcad/choose/job-priority": "Default Priority",
31+
"kubernetes/mcad/choose/scheduler": "My administrator has already installed and configured MCAD"
32+
}
33+
}

Diff for: tests/kind/profiles/non-gpu1/ray-autoscaler.arm64

+31
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,31 @@
1+
{
2+
"name": "ray-autoscaler",
3+
"creationTime": 1660657756574,
4+
"lastModifiedTime": 1660675440396,
5+
"lastUsedTime": 1664202470263,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Ray Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/qiskit\",\"Base image\":\"rayproject/ray:2.3.0-py38-aarch64\",\"Command line prefix\":\"python3 main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": {
18+
"Number of Workers": 1,
19+
"CPUs per worker": "200m",
20+
"GPUs per worker": 0,
21+
"Memory per worker": "1.25Gi",
22+
"Ephemeral Storage per worker": "5Gi"
23+
},
24+
"kubernetes/context": "kind-codeflare-test",
25+
"kubernetes/choose/ns": "default",
26+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
27+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
28+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
29+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Use the Ray Autoscaler"
30+
}
31+
}

Diff for: tests/kind/profiles/non-gpu2/keep-it-simple.arm64

+31
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,31 @@
1+
{
2+
"name": "keep-it-simple",
3+
"creationTime": 1660657756574,
4+
"lastModifiedTime": 1660658307175,
5+
"lastUsedTime": 1678746664684,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Ray Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/qiskit\",\"Base image\":\"rayproject/ray:2.3.0-py38-aarch64\",\"Command line prefix\":\"python3 main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": {
18+
"Number of Workers": 1,
19+
"CPUs per worker": "500m",
20+
"GPUs per worker": 0,
21+
"Memory per worker": "1.5Gi",
22+
"Ephemeral Storage per worker": "5Gi"
23+
},
24+
"kubernetes/context": "kind-codeflare-test",
25+
"kubernetes/choose/ns": "default",
26+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
27+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
28+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
29+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Keep It Simple"
30+
}
31+
}

Diff for: tests/kind/profiles/non-gpu3/keep-it-simple.arm64

+31
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,31 @@
1+
{
2+
"name": "keep-it-simple",
3+
"creationTime": 1660657756574,
4+
"lastModifiedTime": 1660658307175,
5+
"lastUsedTime": 1660742079201,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Ray Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/qiskit\",\"Base image\":\"rayproject/ray:2.3.0-py38-aarch64\",\"Command line prefix\":\"python3 main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": {
18+
"Number of Workers": 1,
19+
"CPUs per worker": "500m",
20+
"GPUs per worker": 0,
21+
"Memory per worker": "1.5Gi",
22+
"Ephemeral Storage per worker": "5Gi"
23+
},
24+
"kubernetes/context": "kind-codeflare-test",
25+
"kubernetes/choose/ns": "default",
26+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
27+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
28+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
29+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Keep It Simple"
30+
}
31+
}

Diff for: tests/kind/profiles/non-gpu4/keep-it-simple.arm64

+31
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,31 @@
1+
{
2+
"name": "keep-it-simple",
3+
"creationTime": 1660657756574,
4+
"lastModifiedTime": 1660658307175,
5+
"lastUsedTime": 1660742079201,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Ray Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/qiskit\",\"Base image\":\"rayproject/ray:2.3.0-py38-aarch64\",\"Command line prefix\":\"python3 main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": {
18+
"Number of Workers": 1,
19+
"CPUs per worker": "500m",
20+
"GPUs per worker": 0,
21+
"Memory per worker": "1.5Gi",
22+
"Ephemeral Storage per worker": "5Gi"
23+
},
24+
"kubernetes/context": "kind-codeflare-test",
25+
"kubernetes/choose/ns": "default",
26+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
27+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
28+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
29+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Keep It Simple"
30+
}
31+
}

Diff for: tests/kind/profiles/non-gpu5/keep-it-simple.arm64

+25
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,25 @@
1+
{
2+
"name": "keep-it-simple",
3+
"creationTime": 1664149787016,
4+
"lastModifiedTime": 1676828268772,
5+
"lastUsedTime": 1678152542012,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Ray Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/qiskit\",\"Base image\":\"rayproject/ray:2.3.0-py38-aarch64\",\"Command line prefix\":\"python3 main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": "{\"Number of Workers\":1,\"CPUs per worker\":\"500m\",\"GPUs per worker\":0,\"Memory per worker\":\"1.5Gi\",\"Ephemeral Storage per worker\":\"5Gi\"}",
18+
"kubernetes/context": "kind-codeflare-test",
19+
"kubernetes/choose/ns": "default",
20+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
21+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
22+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
23+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Keep It Simple"
24+
}
25+
}

Diff for: tests/kind/profiles/non-gpu6/keep-it-simple.arm64

+25
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,25 @@
1+
{
2+
"name": "keep-it-simple",
3+
"creationTime": 1664149787016,
4+
"lastModifiedTime": 1676828268772,
5+
"lastUsedTime": 1678637996635,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Torch Native Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/torchx\",\"Base image\":\"bitnami/pytorch:1.13.1\",\"Command line prefix\":\"python3 compute_world_size/main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": "{\"Number of Workers\":1,\"CPUs per worker\":\"500m\",\"GPUs per worker\":0,\"Memory per worker\":\"1.5Gi\",\"Ephemeral Storage per worker\":\"5Gi\"}",
18+
"kubernetes/context": "kind-codeflare-test",
19+
"kubernetes/choose/ns": "default",
20+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
21+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
22+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
23+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Keep It Simple"
24+
}
25+
}

Diff for: tests/kind/profiles/non-gpu6/mcad-default.arm64

+27
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,27 @@
1+
{
2+
"name": "mcad-default",
3+
"creationTime": 1660657756574,
4+
"lastModifiedTime": 1678638052528,
5+
"lastUsedTime": 1678980273687,
6+
"choices": {
7+
"madwizard/apriori/use-gpu": "don't use gpus",
8+
"madwizard/apriori/arch": "arm64",
9+
"madwizard/apriori/platform": "darwin",
10+
"madwizard/apriori/mac-installer": "Homebrew",
11+
"madwizard/apriori/in-terminal": "HTML",
12+
"ml/codeflare": "Submit a new Run",
13+
"ml/codeflare/run": "Bring Your Own Torch Native Code",
14+
"ml/codeflare/training/byoc/form": "{\"Path to source\":\"$PWD/tests/kind/inputs/torchx\",\"Base image\":\"bitnami/pytorch:1.13.1\",\"Command line prefix\":\"python3 compute_world_size/main.py\"}",
15+
"kubernetes/choose/secret/image-pull": "No secret needed, since my image is public",
16+
"s3/choose/bucket/maybe": "My data is not stored in S3",
17+
"ml/ray/start/resources": "{\"Number of Workers\":1,\"CPUs per worker\":\"500m\",\"GPUs per worker\":0,\"Memory per worker\":\"1.5Gi\",\"Ephemeral Storage per worker\":\"5Gi\"}",
18+
"kubernetes/context": "kind-codeflare-test",
19+
"kubernetes/choose/ns": "default",
20+
"ml/ray/storage/s3/maybe": "My code does not use Ray Workflows",
21+
"ml/ray/cluster/choose": "codeflare-test-ray-cluster",
22+
"ml/ray/cluster/choose/kubernetes": "codeflare-test-ray-cluster",
23+
"ml/ray/cluster/kubernetes/choose-pod-scheduler": "Use the Multi-user Enhanced Kubernetes Scheduler",
24+
"kubernetes/mcad/choose/job-priority": "Default Priority",
25+
"kubernetes/mcad/choose/scheduler": "MCAD with the Default Kubernetes Scheduler"
26+
}
27+
}

Diff for: tests/kind/run.sh

+5-3
Original file line numberDiff line numberDiff line change
@@ -69,7 +69,9 @@ function start_kind {
6969
function run {
7070
local profileFull="$1"
7171
local variant=$(dirname "$profileFull")
72-
local profile=$(basename "$profileFull")
72+
local profileBase=$(basename "$profileFull")
73+
local profile=$(basename "$profileFull").$(uname -m)
74+
7375
export MWPROFILES_PATH="$MWPROFILES_PATH_BASE"/$variant
7476
mkdir -p "$MWPROFILES_PATH"
7577

@@ -95,10 +97,10 @@ function run {
9597
echo "[Test] Run: expecting dashdash: $expectedDashDash"
9698
fi
9799

98-
local PRE="$MWPROFILES_PATH_BASE"/../profiles.d/$profile/pre
100+
local PRE="$MWPROFILES_PATH_BASE"/../profiles.d/$profileBase/pre
99101
if [ -f "$PRE" ]; then
100102
echo "[Test] Running pre guidebooks for profile=$profile"
101-
cat "$PRE" | xargs -n1 "$ROOT"/bin/codeflare -p $profile $yes
103+
cat "$PRE" | xargs -n1 "$ROOT"/bin/codeflare -p $profileBase $yes
102104
fi
103105

104106
echo "[Test] Running with variant=$variant profile=$profile yes=$yes"

0 commit comments

Comments
 (0)