-
Notifications
You must be signed in to change notification settings - Fork 18
194 lines (176 loc) · 7.34 KB
/
adhoc-matrix.yml
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
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
name: Execute ZKVM-Perf (Matrix)
on:
workflow_dispatch:
inputs:
provers:
description: 'Provers to use (comma-separated)'
required: false
type: string
default: 'sp1'
programs:
description: 'Programs to benchmark (comma-separated)'
required: false
type: string
default: 'loop,fibonacci,tendermint,reth1,reth2'
filename:
description: 'Filename for the benchmark'
required: false
type: string
default: 'benchmark'
trials:
description: 'Number of trials to run'
required: false
type: string
default: '1'
sp1_ref:
description: 'SP1 reference (commit hash or branch name)'
required: false
type: string
default: '2e8b0a8'
additional_params:
description: 'Additional parameters as JSON'
required: false
type: string
default: '{"hashfns":"poseidon","shard_sizes":"22"}'
jobs:
run-benchmarks:
strategy:
matrix:
include:
- instance_type: g6.16xlarge
enable_gpu: true
ami_id: ami-079a6a210557ef0e4
- instance_type: r7i.16xlarge
enable_gpu: false
ami_id: ami-079a6a210557ef0e4
name: Run on ${{ matrix.instance_type }}
runs-on: ubuntu-latest
steps:
- name: Configure AWS credentials
uses: aws-actions/configure-aws-credentials@v1
with:
aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
aws-region: ${{ secrets.AWS_REGION }}
- name: Start EC2 runner
id: start-ec2-runner
uses: xJonathanLEI/ec2-github-runner@main
with:
mode: start
# Must use personal access token here as `GITHUB_TOKEN` does not have access to runners.
# Use a fine-grained token with these permissions to at least this repository:
# - Administration: Read and write
# - Contents: Read and write
# - Metadata: Read-only
# - Workflows: Read and write
# - Actions: Read and write
github-token: ${{ secrets.GH_PAT }}
ec2-image-id: ${{ matrix.ami_id }}
ec2-instance-type: ${{ matrix.instance_type }}
subnet-id: ${{ secrets.AWS_SUBNET_ID }}
security-group-id: ${{ secrets.AWS_SG_ID }}
storage-size: 1024
- name: Run benchmarks
id: run-benchmarks
uses: actions/github-script@v6
with:
github-token: ${{ secrets.GH_PAT }}
script: |
const runnerName = '${{ steps.start-ec2-runner.outputs.label }}';
const maxAttempts = 5;
const initialDelay = 30000; // 30 seconds
let triggeredRunId = null;
for (let attempt = 1; attempt <= maxAttempts; attempt++) {
console.log(`Attempt ${attempt} to trigger benchmark workflow`);
await new Promise(resolve => setTimeout(resolve, initialDelay * attempt));
try {
const result = await github.rest.actions.createWorkflowDispatch({
owner: context.repo.owner,
repo: context.repo.repo,
workflow_id: 'run-on-runner.yml',
ref: context.ref,
inputs: {
runner_name: runnerName,
instance_type: '${{ matrix.instance_type }}',
enable_gpu: '${{ matrix.enable_gpu }}',
provers: '${{ inputs.provers }}',
programs: '${{ inputs.programs }}',
filename: '${{ inputs.filename }}_${{ matrix.instance_type }}',
trials: '${{ inputs.trials }}',
sp1_ref: '${{ inputs.sp1_ref }}',
additional_params: '${{ inputs.additional_params }}'
}
});
console.log('Benchmark workflow triggered successfully');
// Wait for the run to appear in the list
for (let i = 0; i < 10; i++) {
await new Promise(resolve => setTimeout(resolve, 5000));
const runs = await github.rest.actions.listWorkflowRuns({
owner: context.repo.owner,
repo: context.repo.repo,
workflow_id: 'run-on-runner.yml',
status: 'in_progress'
});
const recentRun = runs.data.workflow_runs.find(run =>
new Date(run.created_at).getTime() > Date.now() - 60000
);
if (recentRun) {
triggeredRunId = recentRun.id;
break;
}
}
if (triggeredRunId) {
console.log(`Triggered run ID: ${triggeredRunId}`);
break;
} else {
throw new Error('Failed to find the triggered workflow run');
}
} catch (error) {
console.log(`Failed to trigger or find workflow: ${error.message}`);
if (attempt === maxAttempts) {
core.setFailed('Failed to trigger benchmark workflow after multiple attempts');
}
}
}
core.setOutput('triggered-run-id', triggeredRunId);
- name: Wait for benchmark completion
uses: actions/github-script@v6
with:
github-token: ${{ secrets.GH_PAT }}
script: |
const triggeredRunId = ${{ steps.run-benchmarks.outputs.triggered-run-id }};
if (!triggeredRunId) {
core.setFailed('No triggered run ID found');
return;
}
const maxWaitTime = 3600000; // 1 hour in milliseconds
const checkInterval = 60000; // 1 minute in milliseconds
const startTime = Date.now();
while (true) {
const run = await github.rest.actions.getWorkflowRun({
owner: context.repo.owner,
repo: context.repo.repo,
run_id: triggeredRunId
});
if (run.data.status === 'completed') {
console.log(`Benchmark workflow completed with conclusion: ${run.data.conclusion}`);
if (run.data.conclusion !== 'success') {
core.setFailed(`Benchmark workflow failed with conclusion: ${run.data.conclusion}`);
}
break;
}
if (Date.now() - startTime > maxWaitTime) {
core.setFailed('Benchmark workflow did not complete within the maximum wait time');
break;
}
console.log(`Waiting for benchmark to complete... Current status: ${run.data.status}`);
await new Promise(resolve => setTimeout(resolve, checkInterval));
}
- name: Stop EC2 runner
if: always()
uses: xJonathanLEI/ec2-github-runner@main
with:
mode: stop
github-token: ${{ secrets.GH_PAT }}
label: ${{ steps.start-ec2-runner.outputs.label }}
ec2-instance-id: ${{ steps.start-ec2-runner.outputs.ec2-instance-id }}