From 4433015295896548d87a7349606a19cc462e52b1 Mon Sep 17 00:00:00 2001 From: Alex Wu Date: Tue, 3 Nov 2020 19:37:19 -0800 Subject: [PATCH] Release testing things --- doc/dev/release_logs/1.0.0/microbenchmark.txt | 22 ++++++++ doc/dev/release_logs/1.0.1/microbenchmark.txt | 22 ++++++++ release/long_running_tests/actor_deaths.yaml | 52 +++++++++++++++++++ release/long_running_tests/apex.yaml | 52 +++++++++++++++++++ release/long_running_tests/cluster.yaml | 10 ++-- release/long_running_tests/impala.yaml | 52 +++++++++++++++++++ .../long_running_tests/many_actor_tasks.yaml | 52 +++++++++++++++++++ release/long_running_tests/many_drivers.yaml | 52 +++++++++++++++++++ .../many_tasks_serialized_ids.yaml | 52 +++++++++++++++++++ release/long_running_tests/node_failures.yaml | 52 +++++++++++++++++++ release/long_running_tests/pbt.yaml | 52 +++++++++++++++++++ release/long_running_tests/run.sh | 6 ++- release/long_running_tests/serve.yaml | 52 +++++++++++++++++++ release/long_running_tests/serve_failure.yaml | 52 +++++++++++++++++++ release/microbenchmark/cluster.yaml | 10 ++-- release/microbenchmark/run.sh | 4 +- 16 files changed, 581 insertions(+), 13 deletions(-) create mode 100644 doc/dev/release_logs/1.0.0/microbenchmark.txt create mode 100644 doc/dev/release_logs/1.0.1/microbenchmark.txt create mode 100644 release/long_running_tests/actor_deaths.yaml create mode 100644 release/long_running_tests/apex.yaml create mode 100644 release/long_running_tests/impala.yaml create mode 100644 release/long_running_tests/many_actor_tasks.yaml create mode 100644 release/long_running_tests/many_drivers.yaml create mode 100644 release/long_running_tests/many_tasks_serialized_ids.yaml create mode 100644 release/long_running_tests/node_failures.yaml create mode 100644 release/long_running_tests/pbt.yaml mode change 100644 => 100755 release/long_running_tests/run.sh create mode 100644 release/long_running_tests/serve.yaml create mode 100644 release/long_running_tests/serve_failure.yaml diff --git a/doc/dev/release_logs/1.0.0/microbenchmark.txt b/doc/dev/release_logs/1.0.0/microbenchmark.txt new file mode 100644 index 000000000..711cf1c13 --- /dev/null +++ b/doc/dev/release_logs/1.0.0/microbenchmark.txt @@ -0,0 +1,22 @@ +single client get calls per second 30584.85 +- 661.06 +single client put calls per second 23923.57 +- 360.12 +multi client put calls per second 124576.32 +- 1847.41 +single client get calls (Plasma Store) per second 7940.16 +- 456.8 +single client put calls (Plasma Store) per second 4956.74 +- 31.4 +multi client put calls (Plasma Store) per second 7705.71 +- 120.7 +single client put gigabytes per second 13.74 +- 11.88 +multi client put gigabytes per second 36.83 +- 1.89 +single client tasks sync per second 977.6 +- 9.39 +single client tasks async per second 14228.3 +- 195.64 +multi client tasks async per second 36049.11 +- 790.33 +1:1 actor calls sync per second 1454.92 +- 20.51 +1:1 actor calls async per second 6612.21 +- 195.01 +1:1 actor calls concurrent per second 6034.76 +- 141.42 +1:n actor calls async per second 14032.19 +- 434.44 +n:n actor calls async per second 36895.26 +- 642.94 +n:n actor calls with arg async per second 9548.31 +- 218.19 +1:1 async-actor calls sync per second 1031.86 +- 17.59 +1:1 async-actor calls async per second 3675.11 +- 37.81 +1:1 async-actor calls with args async per second 2268.17 +- 20.47 +1:n async-actor calls async per second 11776.12 +- 641.59 +n:n async-actor calls async per second 24358.79 +- 362.87 diff --git a/doc/dev/release_logs/1.0.1/microbenchmark.txt b/doc/dev/release_logs/1.0.1/microbenchmark.txt new file mode 100644 index 000000000..cebc0ba9d --- /dev/null +++ b/doc/dev/release_logs/1.0.1/microbenchmark.txt @@ -0,0 +1,22 @@ +single client get calls per second 30921.88 +- 291.82 +single client put calls per second 26506.75 +- 278.09 +multi client put calls per second 129520.73 +- 2304.11 +single client get calls (Plasma Store) per second 7555.25 +- 109.32 +single client put calls (Plasma Store) per second 4936.57 +- 31.22 +multi client put calls (Plasma Store) per second 7459.17 +- 271.1 +single client put gigabytes per second 11.87 +- 12.03 +multi client put gigabytes per second 36.23 +- 2.17 +single client tasks sync per second 1044.57 +- 16.2 +single client tasks async per second 14319.28 +- 105.0 +multi client tasks async per second 35140.53 +- 498.82 +1:1 actor calls sync per second 1545.69 +- 25.38 +1:1 actor calls async per second 6656.71 +- 268.57 +1:1 actor calls concurrent per second 5851.38 +- 75.39 +1:n actor calls async per second 11856.28 +- 322.15 +n:n actor calls async per second 37089.65 +- 847.78 +n:n actor calls with arg async per second 9919.05 +- 85.45 +1:1 async-actor calls sync per second 1027.93 +- 31.65 +1:1 async-actor calls async per second 3861.91 +- 96.54 +1:1 async-actor calls with args async per second 2351.35 +- 48.13 +1:n async-actor calls async per second 12652.4 +- 637.73 +n:n async-actor calls async per second 24527.67 +- 743.48 diff --git a/release/long_running_tests/actor_deaths.yaml b/release/long_running_tests/actor_deaths.yaml new file mode 100644 index 000000000..4a53fc15a --- /dev/null +++ b/release/long_running_tests/actor_deaths.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-actor-deaths-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/long_running_tests/apex.yaml b/release/long_running_tests/apex.yaml new file mode 100644 index 000000000..6a2853b7f --- /dev/null +++ b/release/long_running_tests/apex.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-apex-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/long_running_tests/cluster.yaml b/release/long_running_tests/cluster.yaml index 074d445ba..76c9995af 100644 --- a/release/long_running_tests/cluster.yaml +++ b/release/long_running_tests/cluster.yaml @@ -1,4 +1,4 @@ -cluster_name: default +cluster_name: long-running-release-tests-alex min_workers: 0 max_workers: 0 target_utilization_fraction: 0.8 @@ -7,14 +7,14 @@ idle_timeout_minutes: 5 # Cloud-provider specific configuration. provider: type: aws - region: us-west-2 - availability_zone: us-west-2a + region: us-west-1 + availability_zone: us-west-1a auth: ssh_user: ubuntu head_node: InstanceType: m5.2xlarge - ImageId: ami-0888a3b5189309429 # DLAMI 7/1/19 + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 BlockDeviceMappings: - DeviceName: /dev/sda1 Ebs: @@ -22,7 +22,7 @@ head_node: worker_nodes: InstanceType: m5.large - ImageId: ami-0888a3b5189309429 # DLAMI 7/1/19 + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 BlockDeviceMappings: - DeviceName: /dev/sda1 Ebs: diff --git a/release/long_running_tests/impala.yaml b/release/long_running_tests/impala.yaml new file mode 100644 index 000000000..6a2853b7f --- /dev/null +++ b/release/long_running_tests/impala.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-apex-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/long_running_tests/many_actor_tasks.yaml b/release/long_running_tests/many_actor_tasks.yaml new file mode 100644 index 000000000..46951d1b8 --- /dev/null +++ b/release/long_running_tests/many_actor_tasks.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-actor-tasks-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/long_running_tests/many_drivers.yaml b/release/long_running_tests/many_drivers.yaml new file mode 100644 index 000000000..95b67e417 --- /dev/null +++ b/release/long_running_tests/many_drivers.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-drivers-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/long_running_tests/many_tasks_serialized_ids.yaml b/release/long_running_tests/many_tasks_serialized_ids.yaml new file mode 100644 index 000000000..f3591ef6d --- /dev/null +++ b/release/long_running_tests/many_tasks_serialized_ids.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-serialized-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/long_running_tests/node_failures.yaml b/release/long_running_tests/node_failures.yaml new file mode 100644 index 000000000..ec3116d31 --- /dev/null +++ b/release/long_running_tests/node_failures.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-node-failures-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/long_running_tests/pbt.yaml b/release/long_running_tests/pbt.yaml new file mode 100644 index 000000000..27cafd61d --- /dev/null +++ b/release/long_running_tests/pbt.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-pbt-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/long_running_tests/run.sh b/release/long_running_tests/run.sh old mode 100644 new mode 100755 index e9288de1c..4386e228f --- a/release/long_running_tests/run.sh +++ b/release/long_running_tests/run.sh @@ -53,7 +53,7 @@ wheel="https://s3-us-west-2.amazonaws.com/ray-wheels/$ray_branch/$commit/ray-$ra echo set-window-option -g mouse on > ~/.tmux.conf echo 'termcapinfo xterm* ti@:te@' > ~/.screenrc # Serve load testing tool -rm -r wrk || true && git clone https://github.com/wg/wrk.git wrk && cd wrk && make -j && sudo cp wrk /usr/local/bin +rm -rf wrk || true && git clone https://github.com/wg/wrk.git wrk && cd wrk && make -j && sudo cp wrk /usr/local/bin pip install -U pip unset RAY_ADDRESS source activate tensorflow_p36 @@ -62,5 +62,7 @@ pip install --upgrade pip pip install -U tensorflow==1.14 pip install -q -U "$wheel" Click pip install -q "ray[all]" "gym[atari]" -python "workloads/$workload.py" +ray stop +sleep 5 +python "$HOME/workloads/$workload.py" diff --git a/release/long_running_tests/serve.yaml b/release/long_running_tests/serve.yaml new file mode 100644 index 000000000..f780c1811 --- /dev/null +++ b/release/long_running_tests/serve.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-serve-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/long_running_tests/serve_failure.yaml b/release/long_running_tests/serve_failure.yaml new file mode 100644 index 000000000..eceafaafa --- /dev/null +++ b/release/long_running_tests/serve_failure.yaml @@ -0,0 +1,52 @@ +cluster_name: long-running-serve-failure-alex +min_workers: 0 +max_workers: 0 +target_utilization_fraction: 0.8 +idle_timeout_minutes: 5 + +# Cloud-provider specific configuration. +provider: + type: aws + region: us-west-1 + availability_zone: us-west-1a +auth: + ssh_user: ubuntu + +head_node: + InstanceType: m5.2xlarge + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + +worker_nodes: + InstanceType: m5.large + ImageId: ami-05fbe06ba1b862ef6 # DLAMI 7/1/19 + BlockDeviceMappings: + - DeviceName: /dev/sda1 + Ebs: + VolumeSize: 150 + + # Run workers on spot by default. Comment this out to use on-demand. + InstanceMarketOptions: + MarketType: spot + +# List of shell commands to run to set up nodes. +setup_commands: [] + +# Custom commands that will be run on the head node after common setup. +head_setup_commands: [] + +# Custom commands that will be run on worker nodes after common setup. +worker_setup_commands: [] + +# Command to start ray on the head node. You don't need to change this. +head_start_ray_commands: [] + +# Command to start ray on worker nodes. You don't need to change this. +worker_start_ray_commands: [] + +file_mounts: + "$HOME/run.sh": "run.sh" + "$HOME/workloads": "workloads" diff --git a/release/microbenchmark/cluster.yaml b/release/microbenchmark/cluster.yaml index 0ee31dbc6..e7860a6a8 100644 --- a/release/microbenchmark/cluster.yaml +++ b/release/microbenchmark/cluster.yaml @@ -1,4 +1,4 @@ -cluster_name: ray-release-microbenchmark +cluster_name: ray-release-microbenchmark-alex min_workers: 0 max_workers: 0 target_utilization_fraction: 0.8 @@ -7,14 +7,14 @@ idle_timeout_minutes: 5 # Cloud-provider specific configuration. provider: type: aws - region: us-west-2 - availability_zone: us-west-2a + region: us-west-1 + availability_zone: us-west-1a auth: ssh_user: ubuntu head_node: InstanceType: m4.16xlarge - ImageId: ami-06d51e91cea0dac8d # Ubuntu 18.04 + ImageId: ami-03fac5402e10ea93b # Ubuntu 18.04 BlockDeviceMappings: - DeviceName: /dev/sda1 Ebs: @@ -22,7 +22,7 @@ head_node: worker_nodes: InstanceType: m5.large - ImageId: ami-06d51e91cea0dac8d # Ubuntu 18.04 + ImageId: ami-03fac5402e10ea93b # Ubuntu 18.04 BlockDeviceMappings: - DeviceName: /dev/sda1 Ebs: diff --git a/release/microbenchmark/run.sh b/release/microbenchmark/run.sh index 7aa2fd09d..ef217177a 100755 --- a/release/microbenchmark/run.sh +++ b/release/microbenchmark/run.sh @@ -47,14 +47,16 @@ echo "commit: $commit" echo "branch: $ray_branch" echo "workload: $workload" -wheel="https://s3-us-west-2.amazonaws.com/ray-wheels/$ray_branch/$commit/ray-$ray_version-cp38-cp38-manylinux1_x86_64.whl" +wheel="https://s3-us-west-2.amazonaws.com/ray-wheels/$ray_branch/$commit/ray-$ray_version-cp36-cp36m-manylinux1_x86_64.whl" echo set-window-option -g mouse on > ~/.tmux.conf echo 'termcapinfo xterm* ti@:te@' > ~/.screenrc +set -x pip uninstall -y -q ray pip install --upgrade pip pip install -U "$wheel" +set +x unset RAY_ADDRESS OMP_NUM_THREADS=64 ray microbenchmark