GPU compute cluster faster on Windows 10 than Ubuntu 18.04 - windows

Info: My windows and Ubuntu system is the same. I just swap out my boot drive using a removable drive bay.
*Not using Docker
CPU: i9-7920X (128 GB DDR4)
GPUs:
2 Titan Vs and
RTX 2070 (Display)
Windows 10:
Nvidia Driver 416.81 in TCC cluster and
Cuda 10
Ubuntu Desktop 18.04:
Nvidia Driver 410.72 (not sure if need to cluster on Linux) and
Cuda 10
Test Software:
Running TensorFlow cifar10 esitimator - same training data - (link at bottom)
python cifar10_main.py --data-dir=/tmp --job-dir=/tmp/cifar10 --num-gpus=2 --train-steps=10000
Windows completes in 420 seconds (7.0 minutes)
Ubuntu completes in 614 seconds (10.2 minutes)
https://github.com/tensorflow/models/tree/master/tutorials/image/cifar10_estimator
Why is Ubuntu so much slower? Do the GPUs need to be clustered on Linux or are the different drivers (Ubuntu vs. Windows) the issue?
Thanks.

Related

How should I install npm, I tried all of the command but none of worked for me

npm install
audited 17250 packages in 7.466s
26 packages are looking for funding
run `npm fund` for details
found 0 vulnerabilities
My System Info:
1.OS Name Microsoft Windows 10 Home Single Language
Version 10.0.18363 Build 18363
Other OS Description Not Available
OS Manufacturer Microsoft Corporation
System Name DESKTOP-7CV9HTK
System Manufacturer HP
System Model HP ENVY Laptop 13-ah1xxx
System Type x64-based PC
System SKU 5HZ05PA#AKL
Processor Intel(R) Core(TM) i7-8565U CPU # 1.80GHz, 1992 Mhz, 4 Core(s), 8 Logical Processor(s)
BIOS Version/Date Insyde F.12, 11/8/2018
SMBIOS Version 3.0
Embedded Controller Version 69.72
BIOS Mode UEFI
BaseBoard Manufacturer HP
BaseBoard Product 850B
BaseBoard Version 69.72
Platform Role Mobile
Secure Boot State On
PCR7 Configuration Elevation Required to View
Windows Directory C:\WINDOWS
System Directory C:\WINDOWS\system32
Boot Device \Device\HarddiskVolume2
Locale United States
Hardware Abstraction Layer Version = "10.0.18362.387"
User Name DESKTOP-7CV9HTK\Hanifullah Jamalzai
Time Zone Afghanistan Standard Time
Installed Physical Memory (RAM) 16.0 GB
Total Physical Memory 15.8 GB
Available Physical Memory 9.89 GB
Total Virtual Memory 18.2 GB
Available Virtual Memory 10.3 GB
Page File Space 2.38 GB
Page File C:\pagefile.sys
Kernel DMA Protection Off
Virtualization-based security Not enabled
Device Encryption Support Elevation Required to View
Hyper-V - VM Monitor Mode Extensions Yes
Hyper-V - Second Level Address Translation Extensions Yes
Hyper-V - Virtualization Enabled in Firmware Yes
Hyper-V - Data Execution Protection Yes
Npm is recognized on cmd but not on vscode?
Assuming you are using windows, you need to update the path in environment/user variables AND restart vscode for it to recognize the changes.
I had Yarn Package Manager which is an alternative to npm. The packages you are trying to install can not be installed with npm, "npm install" will not work because it is not the format used by yarn when you are trying to install some package.

Is it possible to use Docker without Windows 10 pro?

I need to install Docker on my pc with Windows 10 home. I read that I can only install Docker Toolbox. Is there any way to have the latest Docker version instead without upgrading my pc to windows 10 pro?
Thanks
Update
Docker can now be installed on Windows 10 Home (version 2004 or higher).
Refer to this article for installation instructions
https://docs.docker.com/docker-for-windows/install-windows-home/
Old Answer
Docker for Windows requires Hyper-V, and Hyper-V requires Windows 10 Pro (or Windows Server). So no, you can't run Docker without upgrading.
https://docs.docker.com/docker-for-windows/install/
README FIRST for Docker Toolbox and Docker Machine users: Docker for Windows requires Microsoft Hyper-V to run. The Docker for Windows installer enables Hyper-V for you, if needed, and restart your machine.
https://learn.microsoft.com/en-us/virtualization/hyper-v-on-windows/quick-start/enable-hyper-v
Check Requirements
Windows 10 Enterprise, Professional, or Education
64-bit Processor with Second Level Address Translation (SLAT).
CPU support for VM Monitor Mode Extension (VT-c on Intel CPU's).
Minimum of 4 GB memory.
The Hyper-V role cannot be installed on Windows 10 Home.
You can now install Docker Desktop on Windows Home machines using the WSL 2 backend. Docker Desktop on Windows Home is a full version of Docker Desktop for Linux container development.
https://docs.docker.com/docker-for-windows/install-windows-home/
Install Windows 10, version 2004 or higher.
Enable the WSL 2 feature on Windows. For detailed instructions, refer to the Microsoft documentation.
https://learn.microsoft.com/en-us/windows/wsl/install-win10
The following hardware prerequisites are required to successfully run WSL 2 on Windows 10 Home:
64 bit processor with Second Level Address Translation (SLAT)
4GB system RAM
BIOS-level hardware virtualization support must be enabled in the BIOS settings. For more information, see Virtualization.
https://docs.docker.com/docker-for-windows/troubleshoot/#virtualization-must-be-enabled
Download and install the Linux kernel update package.
https://learn.microsoft.com/en-us/windows/wsl/wsl2-kernel

How to install informatica 10.1.0 on windows 10 Pro 64-bit?

How to install informatica 10.1.0 on windows 10 Pro 64 bit?
I have downloaded files from https://edelivery.oracle.com/. Extracted files and double clicked on install.bat file. But only a black screen flashes for a second.
Informatica 10.1 Files
System Details:
Operating system: Windows 10 Pro 64-bit
RAM: 16 GB
Java version: Java 7 64-bit (Also tried with Java 8)
Is there any other prerequisite for informatica 10.1.0?
Is it compatible with Windows 10 Pro 64-bit?
Please give me any solution or suggestion.
Informatica server does not support on windows 10 pro.

Windows Server 2012 R2 in Virtualbox error

I'm trying to install Windows Server 2012 R2 on Virtualbox but I get this error all the time:
Proccessor: Intel(R) Core(TM) i5 CPU
System type: 64-bit
ISO file: en_windows_server_2012_x64_dvd_915478.iso
I've tried many things and had no result, I tried on another computer but got the same error.
When creating a new VirtualMachine I used "Windows > Other Windows" and created a new VDI
Common problem. You have to activate virtualization in your bios. If doesn't matter if you have a 64 bits CPU if you don't have the virtualization activated. This is an example.

how do I get my ubuntu VM to auto adjust the screen resolution to fit my macbook pro or an external monitor?

I'm running an ubuntu guest on a mac osx host using virtual box. I'm trying to get the screen resolution to auto adjust to both my macbook pro as well as an external (2nd) monitor.
The tutorials available on line have me using the CLI to add resolutions like
xrandr --addmode 1440x900
I've already tried:
xrandry --newmode "1440x900_60.00" 106.50 1440 1528 1672 1904 900 903 909 934 -hsync +vsync
so, when I type 'xrandr" this is the output:
Screen 0: minimum 64 x 64, current 1024 x 768, maximum 32000 x 32000
VBOX0 connected 1024x768+0+0 0mm x 0mm
1024x768 60.0*+
800x600 60.0
640x480 60.0
So, how do I get my ubuntu VM to auto adjust the screen resolution to fit my macbook pro or an external monitor?
The key was to increase the Video Memory in VirtualBox for my ubuntu VM. See this link for the answer. See this question on the ubuntu forums.
What worked for me was installing virtualbox-guest-x11.
Host: mid 2014 MacBookPro 15in
Guest: Ubuntu 12.04.5 LTS Desktop AMD64
sudo apt-get install virtualbox-guest-x11
I have be struggling to fix this for some time now. I am using Mac OS Catalina 10.15.3 and Trying to install Ubuntu 16.04 on Virtual Box version 6.0.14.
I had tried adding 'Guest Additions CD' image, installing virtualbox-guest-x11 etc but didn't work. At last what worked for me was this
Created a new VM then before installing OS I changed below configuration
Virtual Box > Settings > Display
Increased Video Memory from default 16Mb to 128 Mb and graphics controller to 'VBoxSVGA'
Then installed Ubuntu 16.04, and by default the resolution is perfect.

Resources