@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 8b82a89997
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 16, 23:15 UTC

[ci] Add AMD, Neuron, Intel tests for AWS CI and turn off default soft fail for GPU tests (#5464)

Signed-off-by: kevin <kevin@anyscale.com>

Kevin H. Luu committed Jun 12, 2024 at 14:00 UTC 8b82a89997826af8e0e4ecfaaed60f3b28b1baed
1 file changed +29 -1
.buildkite/test-template-aws.j2
+29 -1
@@ -19,6 +19,34 @@ steps:
19 limit: 5
20 - wait
21
22 + - group: "AMD Tests"
23 + depends_on: ~
24 + steps:
25 + {% for step in steps %}
26 + {% if step.mirror_hardwares and "amd" in step.mirror_hardwares %}
27 + - label: "AMD: {{ step.label }}"
28 + agents:
29 + queue: amd
30 + command: bash .buildkite/run-amd-test.sh "cd {{ (step.working_dir or default_working_dir) | safe }} ; {{ step.command or (step.commands | join(" ; ")) | safe }}"
31 + env:
32 + DOCKER_BUILDKIT: "1"
33 + soft_fail: true
34 + {% endif %}
35 + {% endfor %}
36 +
37 + - label: "Neuron Test"
38 + depends_on: ~
39 + agents:
40 + queue: neuron
41 + command: bash .buildkite/run-neuron-test.sh
42 + soft_fail: false
43 +
44 + - label: "Intel Test"
45 + depends_on: ~
46 + agents:
47 + queue: intel
48 + command: bash .buildkite/run-cpu-test.sh
49 +
50 {% for step in steps %}
51 - label: "{{ step.label }}"
52 agents:
@@ -31,7 +59,7 @@ steps:
59 {% else %}
60 queue: gpu_1_queue
61 {% endif %}
34 - soft_fail: true
62 + soft_fail: {{ step.soft_fail or false }}
63 {% if step.parallelism %}
64 parallelism: {{ step.parallelism }}
65 {% endif %}