hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-06 10:31:39 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions
ray/release/benchmarks/scheduling.yaml

27 lines
576 B
YAML
Raw Normal View History

Migrate scalability tests (#22901) This PR migrates scalability tests to the new infra. I had to copy the benchmarks folder to the release folder to make it work. I will remove some unnecessary files (e.g., benchmark.yaml or wait_for_cluster file) Alternatively we can support a different path than /release from the tool, but I think this way is cleaner. I am open to suggestion though cc @krfricke
2022-03-09 10:22:41 +09:00
cloud_id: {{env["ANYSCALE_CLOUD_ID"]}}
region: us-west-2
max_workers: 999
head_node_type:
name: head_node
use smaller instance for scheduling tests (#25635) m5.16xlarge instances have 64 CPU and 256GB memory, which are overkill for scheduling tests that do not have a lot of computations. Use smaller instance m5.4xlarge to save cost and make allocating instances easier.
2022-06-10 10:09:35 -07:00
instance_type: m5.4xlarge
Migrate scalability tests (#22901) This PR migrates scalability tests to the new infra. I had to copy the benchmarks folder to the release folder to make it work. I will remove some unnecessary files (e.g., benchmark.yaml or wait_for_cluster file) Alternatively we can support a different path than /release from the tool, but I think this way is cleaner. I am open to suggestion though cc @krfricke
2022-03-09 10:22:41 +09:00
resources:
use smaller instance for scheduling tests (#25635) m5.16xlarge instances have 64 CPU and 256GB memory, which are overkill for scheduling tests that do not have a lot of computations. Use smaller instance m5.4xlarge to save cost and make allocating instances easier.
2022-06-10 10:09:35 -07:00
# Assume the node has 64 CPU instead of 16.
# This should be fine since each task has little
# computation in scheduling tests.
Migrate scalability tests (#22901) This PR migrates scalability tests to the new infra. I had to copy the benchmarks folder to the release folder to make it work. I will remove some unnecessary files (e.g., benchmark.yaml or wait_for_cluster file) Alternatively we can support a different path than /release from the tool, but I think this way is cleaner. I am open to suggestion though cc @krfricke
2022-03-09 10:22:41 +09:00
cpu: 64
custom_resources:
node: 1
worker_node_types:
- name: worker_node
use smaller instance for scheduling tests (#25635) m5.16xlarge instances have 64 CPU and 256GB memory, which are overkill for scheduling tests that do not have a lot of computations. Use smaller instance m5.4xlarge to save cost and make allocating instances easier.
2022-06-10 10:09:35 -07:00
instance_type: m5.4xlarge
Migrate scalability tests (#22901) This PR migrates scalability tests to the new infra. I had to copy the benchmarks folder to the release folder to make it work. I will remove some unnecessary files (e.g., benchmark.yaml or wait_for_cluster file) Alternatively we can support a different path than /release from the tool, but I think this way is cleaner. I am open to suggestion though cc @krfricke
2022-03-09 10:22:41 +09:00
min_workers: 31
max_workers: 31
use_spot: false
resources:
cpu: 64
custom_resources:
node: 1
Reference in a new issue Copy permalink