Amazon ec2 Ray不分布在群集中的所有可用CPU上
我的问题是ray不会分配给我的员工 我总共有16个内核,因为ubuntu的每个ec2 aws实例上都有8个CPU 然而,当我启动我的ray cluster并提交python脚本时,它只分布在8个内核上,因为只有8个PID显示出可用性 值得注意的是,我无法访问EC2实例上的ray dashboard,我只能通过打印正在使用的PID来获得这些信息 如何让我的脚本由所有16个CPU进行UT化,从而显示16PID用于执行脚本 这是我的剧本:Amazon ec2 Ray不分布在群集中的所有可用CPU上,amazon-ec2,multiprocessing,cluster-computing,distributed-computing,ray,Amazon Ec2,Multiprocessing,Cluster Computing,Distributed Computing,Ray,我的问题是ray不会分配给我的员工 我总共有16个内核,因为ubuntu的每个ec2 aws实例上都有8个CPU 然而,当我启动我的ray cluster并提交python脚本时,它只分布在8个内核上,因为只有8个PID显示出可用性 值得注意的是,我无法访问EC2实例上的ray dashboard,我只能通过打印正在使用的PID来获得这些信息 如何让我的脚本由所有16个CPU进行UT化,从而显示16PID用于执行脚本 这是我的剧本: import os import ray import tim
import os
import ray
import time
import xgboost
from xgboost.sklearn import XGBClassifier
def printer():
print("INSIDE WORKER " + str(time.time()) +" PID : "+ str(os.getpid()))
# decorators allow for futures to be created for parallelization
@ray.remote
def func_1():
#model = XGBClassifier()
count = 0
for i in range(100000000):
count += 1
printer()
return count
@ray.remote
def func_2():
#model = XGBClassifier()
count = 0
for i in range(100000000):
count += 1
printer()
return count
@ray.remote
def func_3():
count = 0
for i in range(100000000):
count += 1
printer()
return count
def main():
#model = XGBClassifier()
start = time.time()
results = []
ray.init(address='auto')
#append fuction futures
for i in range(10):
results.append(func_1.remote())
results.append(func_2.remote())
results.append(func_3.remote())
#run in parrallel and get aggregated list
a = ray.get(results)
b = 0
#add all values in list together
for j in range(len(a)):
b += a[j]
print(b)
#time to complete
end = time.time()
print(end - start)
if __name__ == '__main__':
main()
这是我的配置:
# A unique identifier for the head node and workers of this cluster.
cluster_name: basic-ray-123454
# The maximum number of workers nodes to launch in addition to the head
# node. This takes precedence over min_workers. min_workers defaults to 0.
max_workers: 2 # this means zero workers
min_workers: 2 # this means zero workers
# Cloud-provider specific configuration.
provider:
type: aws
region: eu-west-2
availability_zone: eu-west-2a
file_mounts_sync_continuously: False
auth:
ssh_user: ubuntu
ssh_private_key: /home/user/.ssh/aws_ubuntu_test.pem
head_node:
InstanceType: c5.2xlarge
ImageId: ami-xxxxxxa6b31fd2c
KeyName: aws_ubuntu_test
BlockDeviceMappings:
- DeviceName: /dev/sda1
Ebs:
VolumeSize: 200
worker_nodes:
InstanceType: c5.2xlarge
ImageId: ami-xxxxx26a6b31fd2c
KeyName: aws_ubuntu_test
file_mounts: {
"/home/ubuntu": "/home/user/RAY_AWS_DOCKER/ray_example_2_4/conda_env.yaml"
}
setup_commands:
- echo "start initialization_commands"
- sudo apt-get update
- sudo apt-get upgrade
- sudo apt-get install -y python-setuptools
- sudo apt-get install -y build-essential curl unzip psmisc
- pip install --upgrade pip
- pip install ray[all]
- echo "all files :"
- ls
# - conda install -c conda-forge xgboost
head_start_ray_commands:
- ray stop
- ulimit -n 65536; ray start --head --port=6379 --object-manager-port=8076 --autoscaling-config=~/ray_bootstrap_config.yaml
worker_start_ray_commands:
- ray stop
- ulimit -n 65536; ray start --address=$RAY_HEAD_IP:6379 --object-manager-port=8076
你能试试ray.get_available_resources()并确保你的集群实际上有16个CPU吗?另外,您是否可以尝试在每个任务中添加print(os.get_pid())以确保只有8个任务同时运行?您是否可以尝试ray.get_available_resources()并确保您的集群实际有16个CPU?另外,您是否可以尝试在每个任务中添加print(os.get_pid()),以确保只有8个任务同时运行?