Google Diet Colab Notebook

[ric]

Originally Posted by Uncwilly

Did GPU72 auto submit it for you?
I checked all the exponents in the file that you posted and it looks like you got credit for each. Look at your account and see which machine is listed.

Manual submission via spider, so their value got posted to "Manual testing". However, I double checked today's submissions, and realized that I had a former submission around noon time, of the very same results.

Hence, the "not needed". Should have force-refreshed results page...

Thus, facepalm on me, and my apologies for the noise.

[Dylan14]

Introducing tf1G.py v0.14:

New feature: you can set the maximum number of loops to run by setting the variable MAXLOOPS. Setting it to 0 will mean the script runs until something happens that causes it to end (like Ctrl+C, updates, etc.).

[chalsall]

It's been a little while since anyone posted here, so...

It empirically appears that Colab is once again viable for both GPU and CPU compute. Several GPU72_TF Notebook users have been getting a generous amount of compute the last month.

Thanks, Google!!!

You /have/ to be aware of what we're doing. Perhaps not the use cases originally envisioned, it can easily be argued that a great deal of education is going on here! I myself have learned a tonne!

One data point for anyone doing work with these things, be aware of something important which just bit me /hard/...

If you launch a script from a Notebook Section's Python code block, and that script then launches a background process, it is NOT killed (read: SIGINT, SIGHUP, SIGKILL) if the section is "Interrupted". Something as simple as:

$Rslt = `./payload.pl </dev/null >/dev/null 2>/dev/null &`;

...in Perl means payload.pl will run until the instance is actually terminated, NOT when the GUI "Section" is interrupted. IPC is needed to deal with this.

Edit: Oh, forgot to mention... A new GPU72_TF Colab user happened by yesterday, and spun up a session in minutes. Cool.

However, he mentioned that he also tried running it on Kaggle, and while it ran fine, his account was banned within minutes. He said he didn't care, as he never used Kaggle anymore. But I thought that worth mentioning. I'll have to put warning language on the GPU72 Notebook code pages.

[ixfd64]

Originally Posted by SimonVall00

I have succesfully run mfaktc in Google Colaboratory. I have created a Github repository with the Notebook i used.



Here is a link to the repository.
https://github.com/SimonVall00/mfaktc-google-colab

This is now returning a 404 error.

[kriesel]

Originally Posted by ixfd64

This is now returning a 404 error.

https://www.mersenneforum.org/showthread.php?t=24839

[petrw1]

Originally Posted by Fan Ming

Compiled new version of gpuowl for Google colab.
ipynb:

import os.path
from google.colab import drive

if not os.path.exists('/content/drive/My Drive'):
  drive.mount('/content/drive')

%cd '/content/drive/My Drive/gpuowl-master/'

!cp 'gpuowl.exe' /usr/local/bin/
!chmod 755 '/usr/local/bin/gpuowl.exe'

!/usr/local/bin/gpuowl.exe -use ORIG_X2,MERGED_MIDDLE

replace the "gpuowl-master/" with is the actual directory the binary located in Google Drive.

What am I doing wrong?
I see a directory called: My Drive/Colab Notebooks
In here are all my workbooks
And here is where I placed gpuowl.exe

Then.....

import os.path
from google.colab import drive

if not os.path.exists('/content/drive/My Drive'):
  drive.mount('/content/drive')

%cd '/content/drive/My Drive/Colab Notebooks/'

!cp 'gpuowl.exe' /usr/local/bin/
!chmod 755 '/usr/local/bin/gpuowl.exe'

!/usr/local/bin/gpuowl.exe -use ORIG_X2,MERGED_MIDDLE

[Errno 2] No such file or directory: '/content/drive/My Drive/Colab Notebooks/'
/content
cp: cannot stat 'gpuowl.exe': No such file or directory
chmod: cannot access '/usr/local/bin/gpuowl.exe': No such file or directory
/bin/bash: /usr/local/bin/gpuowl.exe: No such file or directory

from google.colab import drive
drive.mount('/content/drive')
!chmod 777 '/content/drive/My Drive/gpuowl'
!cd '/content/drive/My Drive/gpuowl' && LD_LIBRARY_PATH="lib:${LD_LIBRARY_PATH}" && chmod 777 gpuowl && chmod 777 worktodo.txt && ./gpuowl -use ORIG_X2 -block 200 -log 120000 -maxAlloc 10240 -user petrw1 -cpu colab/K80

Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True).
/bin/bash: ./gpuowl: Is a directory

I also tried the name I see in the URL window:

[CODE]https://drive.google.com/drive/folders/1DM******************

as

%cd '/content/drive/folders/1DM**********'

Same result.

Thanks

[Fan Ming]

Originally Posted by petrw1

What am I doing wrong?
I see a directory called: My Drive/Colab Notebooks
In here are all my workbooks
And here is where I placed gpuowl.exe

Then.....
[CODE]import os.path
from google.colab import drive

if not os.path.exists('/content/drive/My Drive'):
drive.mount('/content/drive')

%cd '/content/drive/My Drive/Colab Notebooks/'

Try creating another folder (for example, foo) under google drive, and copy gpuowl.exe(without .exe it sometimes failed to recognize it as a binary file, just like the second error /bin/bash: ./gpuowl: Is a directory you got) to that folder, then change directory to '/content/drive/My Drive/foo/'.
Maybe because Colab Notebooks is a special folder, thus caused the error.

[petrw1]

If I go to Google Drive and open directory My Drive.
It shows up on the URL bar as: https://drive.google.com/drive/my-drive
But in the folder list as: My Drive
My Drive contains a file with the source code for each Notebook session I am running and gpuowl.exe.
It has subfolders of
- Colab Notebooks with more source code files of Notebook session I have run
- ColabWork (A new folder I created) with another copy of gpuowl.exe

HOWEVER in my Python window when I run this code:

import os.path
from google.colab import drive

if not os.path.exists('/content/drive/My Drive'):
  drive.mount('/content/drive')

%cd '/content/drive/My Drive'
for x in os.listdir('.'):
    print (x)

I get this:

/content/drive/My Drive
cudapm1
gpuowl

I suspect these were created form WorkBooks I ran recently to install cudaPm1 and GpuOwl.

[petrw1]

I ran the install script provided by kriesel and ended up with this folder/contents:

/content/drive/My Drive/cudapm1
cudapm1-0.20
cudapm1-0.20.tar.gz

/content/drive/My Drive/cudapm1/cudapm1-0.20
readme.txt
CUDAPm1.ini
libcufft.so.5.5
CUDAPm1
cudapm1-setup.txt

readme.txt says to run this as a test from my Python3 workbook:
I read that I might need run in from of this. I tried both.

./CUDAPm1 61408363 -b1 600000 -b2 12000000 -f 3360k

In both cases I get nothing more than:

File "<ipython-input-22-2b58d7006908>", line 18
    run ./CUDAPm1 -cufftbench 1k to 65536k 2 times
         ^
SyntaxError: invalid syntax

There was also an implication I might need to "make" first.
If so am I on the right track here:
https://pypi.org/project/make/

Thanks

[petrw1]

import subprocess

subprocess.call("'/content/drive/My Drive/cudapm1/cudapm1-0.20/CUDAPm1 61408363 -b1 600000 -b2 12000000 -f 3360k'")

FileNotFoundError Traceback (most recent call last)
<ipython-input-15-1b23572579e6> in <module>()
18 import subprocess
19
---> 20 subprocess.call("'/content/drive/My Drive/cudapm1/cudapm1-0.20/CUDAPm1 61408363 -b1 600000 -b2 12000000 -f 3360k'")

/usr/lib/python3.6/subprocess.py in _execute_child(self, args, executable, preexec_fn, close_fds, pass_fds, cwd, env, startupinfo, creationflags, shell, p2cread, p2cwrite, c2pread, c2pwrite, errread, errwrite, restore_signals, start_new_session)
1362 if errno_num == errno.ENOENT:
1363 err_msg += ': ' + repr(err_filename)
-> 1364 raise child_exception_type(errno_num, err_msg, err_filename)
1365 raise child_exception_type(err_msg)
1366

FileNotFoundError: [Errno 2] No such file or directory: "'/content/drive/My Drive/cudapm1/cudapm1-0.20/CUDAPm1 61408363 -b1 600000 -b2 12000000 -f 3360k'": "'/content/drive/My Drive/cudapm1/cudapm1-0.20/CUDAPm1 61408363 -b1 600000 -b2 12000000 -f 3360k'"

[chalsall]

Originally Posted by petrw1

FileNotFoundError: [Errno 2] No such file or directory: "'/content/drive/My Drive/cudapm1/cudapm1-0.20/CUDAPm1 61408363 -b1 600000 -b2 12000000 -f 3360k'": "'/content/drive/My Drive/cudapm1/cudapm1-0.20/CUDAPm1 61408363 -b1 600000 -b2 12000000 -f 3360k'"

Taking a quick stab in the dark... Your Subprocess() call appears to be passing all the parameters to CUDAPm1 as the file name.

Also, wrt your previous message, was your attempt to run "./CUDAPm1" within a Notebook section? If so, you need to do "!./CUDAPm1" instead, since it's within the Python shell.