Help us understand the problem. What is going on with this article?

Google Cloud ML Engineのメモリを調べてみた

More than 1 year has passed since last update.

Google Cloud ML Engineのメモリを調べてみた

https://cloud.google.com/ml-engine/docs/concepts/training-overview

Even though the exact specifications of the machine types are subject to change at any time, you can compare them in terms of relative capability. The following table uses rough "t-shirt" sizing to describe the machine types.

スクリーンショット 2017-09-08 18.47.20.png

ココらへんがいまいち曖昧なので調べてみた

調べ方

import os

mem_bytes = os.sysconf('SC_PAGE_SIZE') * os.sysconf('SC_PHYS_PAGES')
mem_gib = mem_bytes/(1024.**3)
print(str(mem_gib)+"GB")

# another
meminfo = dict((i.split()[0].rstrip(':'),int(i.split()[1])) for i in open('/proc/meminfo').readlines())
mem_gib = meminfo['MemTotal']/(1024.**2)
print(str(mem_gib)+"GB")

Results

Scale tier

BASIC

  • single
    • 14.6924285889 GB

STANDARD_1

  • single
    • 7.01322555542 GB
  • master
    • 7.01322555542 GB
  • ps
    • 14.6924285889 GB
  • worker
    • 7.01322555542 GB

PREMIUM_1

  • single
    • 14.0992164612 GB
  • master
    • 14.0992164612 GB
  • ps
    • 51.1138572693 GB
  • worker
    • 14.0992164612 GB

BASIC_GPU

  • single
    • 29.4575538635 GB

CUSTOM

Machine type

standard

  • 14.6924285889 GB

large_model

  • 51.1138572693 GB

complex_model_s

  • 7.01327514648 GB

complex_model_m

  • 14.0991668701 GB

complex_model_l

  • 28.272064209 GB

standard_gpu

  • 29.4575538635 GB

complex_model_m_gpu

  • 118.048339844 GB

complex_model_l_gpu

  • 118.048339844 GB

コードはこちら

https://github.com/mainyaa/ml-engine-memory-cat

Why do not you register as a user and use Qiita more conveniently?
  1. We will deliver articles that match you
    By following users and tags, you can catch up information on technical fields that you are interested in as a whole
  2. you can read useful information later efficiently
    By "stocking" the articles you like, you can search right away
Comments
Sign up for free and join this conversation.
If you already have a Qiita account
Why do not you register as a user and use Qiita more conveniently?
You need to log in to use this function. Qiita can be used more conveniently after logging in.
You seem to be reading articles frequently this month. Qiita can be used more conveniently after logging in.
  1. We will deliver articles that match you
    By following users and tags, you can catch up information on technical fields that you are interested in as a whole
  2. you can read useful information later efficiently
    By "stocking" the articles you like, you can search right away