Skip to content

Commit

Permalink
[Release tests] Create compute config for new dataset shuffle tests (r…
Browse files Browse the repository at this point in the history
…ay-project#24239)

Use a separate compute config that uses smaller instance types and no object store memory limit for the new shuffle implementation. I verified that the config works on master for dataset_shuffle_* tests.

Related issue number

ray-project#24176: the added tests would verify the instance types which support the new shuffle implementations.
  • Loading branch information
mwtian authored Apr 27, 2022
1 parent 6560a2b commit afdfd20
Show file tree
Hide file tree
Showing 2 changed files with 65 additions and 2 deletions.
Original file line number Diff line number Diff line change
@@ -0,0 +1,19 @@
cloud_id: {{env["ANYSCALE_CLOUD_ID"]}}
region: us-west-2

aws:
BlockDeviceMappings:
- DeviceName: /dev/sda1
Ebs:
VolumeSize: 500

head_node_type:
name: head_node
instance_type: m5.4xlarge

worker_node_types:
- name: worker_node
instance_type: m5.4xlarge
min_workers: 19
max_workers: 19
use_spot: false
48 changes: 46 additions & 2 deletions release/release_tests.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3397,11 +3397,9 @@
run:
timeout: 7200
script: python dataset/sort.py --num-partitions=1000 --partition-size=1e9 --shuffle

wait_for_nodes:
num_nodes: 20
timeout: 900

type: sdk_command
file_manager: sdk

Expand All @@ -3423,11 +3421,57 @@
run:
timeout: 7200
script: python dataset/sort.py --num-partitions=1000 --partition-size=1e9
wait_for_nodes:
num_nodes: 20
timeout: 900
type: sdk_command
file_manager: sdk

- name: dataset_shuffle_random_shuffle_1tb_small_instances
group: core-dataset-tests
working_dir: nightly_tests
legacy:
test_name: dataset_shuffle_random_shuffle_1tb_small_instances
test_suite: dataset_test

stable: false

frequency: nightly
team: core
cluster:
cluster_env: shuffle/shuffle_app_config.yaml
cluster_compute: shuffle/datasets_large_scale_compute_small_instances.yaml

run:
timeout: 7200
script: python dataset/sort.py --num-partitions=1000 --partition-size=1e9 --shuffle
wait_for_nodes:
num_nodes: 20
timeout: 900
type: sdk_command
file_manager: sdk

- name: dataset_shuffle_sort_1tb_small_instances
group: core-dataset-tests
working_dir: nightly_tests
legacy:
test_name: dataset_shuffle_sort_1tb_small_instances
test_suite: dataset_test

stable: false

frequency: nightly
team: core
cluster:
cluster_env: shuffle/shuffle_app_config.yaml
cluster_compute: shuffle/datasets_large_scale_compute_small_instances.yaml

run:
timeout: 7200
script: python dataset/sort.py --num-partitions=1000 --partition-size=1e9
wait_for_nodes:
num_nodes: 20
timeout: 900
type: sdk_command
file_manager: sdk

Expand Down

0 comments on commit afdfd20

Please sign in to comment.