Search the Community

Asking here as nobody can seem to figure this out. I have an application that requires the smallest boot-time/TTL possible with GPUs attached to a VM in GCP CE. To keep cost down, my infrastructure is dependent on starting and stopping dedicated instances as demand increases/decreases. (I would convert to containers but gpu costs are too high) I have achieved sub-5second start times with custom images without GPUs, but as soon as I attach a GPU, the time to "RUNNING" is always past 20-30s. I have tried multiple different distros, clear linux, prepackaged Nvidia driver images, minimal installs of Fedora, minimalised Debian, reductions to kernel and userspace - systemd-analyze says my boot-time is 3s, but starting the VM with a GPU takes 20-30s in "STAGING" before running. This only occurs when the gpu is attached to the VM and when removed the VM starts within the time mentioned by systemd-analyze. It is consistent across all distros and bootimages. Is there any packages or documentation I am missing to speed up this staging-time with a GPU attached or is this a limitation with GCP's internal staging of GPU instances? I'd much appreciate any help or advice.

First of all if I'm posting in the wrong Threads, please guide me as I couldn't find another for hosting advice. I'm trying to start a blog, I've looked at shared and VPS hosting providers, but as I'm a student I have access to AWS free tier as well as Google Cloud. I thinking of saving some money and using them to host my blog. I know I can install WordPress directly, but I would much prefer having a hosting like panel to manage and monitor things, install SSL, create emails for the domain, have a phpMyAdmin or similar a filemanager, etc. I read quite a bit on this and found out that I can use CyberPanel with OpenLiteSpeed Cache or Ajenti for doing it. I also looked at plesk obsidian(free web min version). But I don't really have any first hand experience with these and I have a few questions. 1. Will the panel I choose affect the speed of the website. Is cyberpanel better than ajenti or plesk because it offers OLS? 2. Will the distro I choose (ubuntu/centOS ) have an impact on the speed. 3. are these my only options or is there any better/faster panel that I can/should use? 4. which one of these 3 would you suggest keeping in mind that although I'm a total beginner I'm ready to learn if needed?

Which is the cheapest way to have a personal Linux server.. is it the cloud or raspberry pi. I tried the free trail of Google cloud. It is a bit complicated for Google VM to ssh. Also I don't think i can afford Google VM after the trail period. Don't have that much money. Any ideas? Thanks Bob

Just yesterday (or 10/5/17 if you're a latecomer), Google announced that it is now allowing VM configurations with up to 96 vCPUs and 624GB of memory for its own Compute Engine machine types on their cloud platform. This is a pretty sizable increase over what they offered in March, with VM configs of 64 cores and 465GB of ram. (3) And if the previous options are still not enough and you happen to have workloads compatible with what is mentioned above, you now have even greater options to use your compute budget. Well, you have a beta to try it out and there are even greater offerings coming someday, somewhere... Update 1: Google Cloud Compute will allow you to sign up for a one year trial with $300 of credit (Annual not monthly) if you want to play around try it out. The downside is that during the trial, you can only make VMs with up to 8 cores and 52GB of ram. This configuration is estimated at $345.73/month if that gives you a better indicator of how much the full fat 96 core will cost. What are your thoughts on this and will you end up using it? Sources: (Google) (Tech Crunch 1) (Tech Crunch 2) This is my first tech news and reviews topic so if there are mistakes, please do say so. This post is also compliant with night theme.

!!!GO HERE FOR THE UPDATED VERSION!!!! Those who fist want some numbers in order to see of its worth it-> it IS check my latest Post here! Basics and Introduction With the Folding month going on and competition being massive I decided to think about ways of maxing my PPD besides upgrading and OCing my own Hardware (wich I also do) I am using Google Cloud sometimes for my work as it includes MUCH Data science etc. Setting up an account there and getting a Debian (Linux) VM going with F@H at full blast is pretty easy and can be done in like 30min. I know that with sharing this Ill loose the advantage, BUT competition is nice... BUT the cause is more important! I think and hope that's why we are participating in the F@H month anyway. Basic Infos: Google clouds virtual machines are payed by usage time. Sining up for the free trial will give you 365 days of time and a free budget of 300$ to spend on computing time. After this your Card WONT be charged. You will get notified and if you do not do anything the machines will simply shut down. Therefore this is pretty save concerning the $$$ :). In this guide we will create some machines wich max out the power but burn through the 300€ relative slowly. Therefore this is a nice easy and free way of boosting your PPD ! RIIIGHT so letzzzze go MARIO (The screenshots will be German, because I didn’t want to reset my whole google accounts language, but ill mark + tanslate everything so its easy to follow) 1. You need a Google Account. You can use your existing one or create a new one (multiple are also possible, but I am NOT responsible for anything :P) https://cloud.google.com/ sing in, give them your Card (there they write about not charging you when you’re done with the free 300$ budget. If you have google pay you can use that). Watch out to create a PERSONAL account!! Once your in at the dashboard: Click on compute engine, then you have to wait a few minutes for google cloud to activate some stuff... (see screenshot, everything will be in spoilers Create a new VM Instance Google Cloud free is limited to 12 CPU cores per user. Therefore we have to create two machines. One with 8 and one with 4 cores. Creating one with 16 will result in an „over-quota“ error. Now we will create the 8 core machine, but the process for the 4core guy is the same, I´ll come back to that Leave every setting as is unless I tell you to change it We need a computing optimized machine. The cheapest ones are available in Region us-central and then on Zone us-central-c (-b should work to, but I haven’t tried, price is the same) Choose computing optimized Machine type: c2-standard-8 (I know the others are BA, but the free plan wont allow them ;( LEAVE everything else as is! Click Create on the bottom of the page. Back at the „VM-Instances“ view: wait a minute for your VM to be initialized and booted Info: You can connect to the VM via SSH, so you Putty guys: go! But for our purpose the simplest way is to connect with the build in tools. Click on SSH and wait (copy and paste sometimes only works on Chrome… who would have thought xD. Also install this plugin "SSL for google Cloud" to be able to copy and paste in the browser SSL client https://chrome.google.com/webstore/detail/ssh-for-google-cloud-plat/ojilllmhjhibplnppnamldakhpmdnibd ). Plz do this or you will have to type some long links and your passkey for F@H :P) EDIT: Now it works even with Safari on my Mac... guess you should try and then see if you need Chrome/the Plugin Now you’re in Debian (terminal Version). No need to configure anything, we directly download FAH! To be able to copy and paste go to settings at the „Clipboard settings“ and activate „copy with ctrl…“ Now we basically only need the FAHClient (FAHControl is not needed and unless since we don’t have graphical output :P) Everything is taken from: https://foldingathome.org/support/faq/installation-guides/linux/manual-installation-advanced/ First run (input it with copy and paste) wget https://download.foldingathome.org/releases/public/release/fahclient/debian-testing-64bit/v7.4/fahclient_7.4.4_amd64.deb Thats downloading the client, then sudo dpkg -i --force-depends fahclient_7.4.4_amd64.deb To install FAH A configuration will pop up and ask for Username and Team Nr. -> you know what to do 223518 FTW! ATTENTION HERE: Coppy+Pasting my passkey did NOT work for me! It showed only blank space but with he cursor at the end of the line. This does NOT work! Sadly you have to type it… and plz. Tripple check it!!!! EDDIT: On some browsers and machines it works, on some not... well just check the the key is properly inserted before clicking ok. Set that baby to full thrust! DISABLE autostart!!!! You will get some errors now, but you can ignore them Being back at the command prompt: Optional: The F@H installer complains about the dependency "bzip2". For me everything runs somooth without it, but if you want to install it to be super save: Very simple; its only entering sudo apt install bzip2 20. Normally the Folding process will stop when you disengage the SSH Connection To prevent this we will run screen. Thats sort of a terminal inside the terminal. It will persist even when logged out of SSH. screen 21. To run F@H Input: FAHClient You will get even more error annnnd were done installing! 22. BUT DONT CLOSE THE WINDOW NOW! Disengage form the screen with pressing ctrl+A+D. This will bring you back to the normal Terminal. to see your screen sessions type screen -ls That should show one session with the number 1. Thats where our folding is happening to reconnect screen -r numberOfTheSession to exit form the MAIN terminal AFTER exiting screen exit To create a machine with the 4 cores you have left over: Click on the + thingy on the top of the overview and repeat form point 5, but now with a "c2-standard-4" configuration. The rest of the installation is exactly the same! Anny suggestions, feedback or things I can improve in this tutorial? Please let me know! A HUGE THANKS TO: @Gorgon and @proptiedwho helped me figure some stuf out and contributed some tricks and commands to this guide! -To do: Remote Monitoring; exact configuration for Linux Servers... I have to get it working for myself first...- I want more PPD! (be careful to watch your free balance!) Useful Tips and Tricks Remote Monitoring by @Gorgon How to monitor if everything is running smoothly. Wait 5min, go to the VM-Instances box, click Infos and Monitoring, check if the CPU is pegged. If it is: Congrats! Its running! Should look like this: I have noticed this to update VERRRRy slowly and be kind of inaccurate. The direct way of doing things is to SSH back in the server and running top thats a "taskmanager" to get out: Ctrl+C How to check the log files once the SSH connection was closed. Log in to google cloud on the left side of your dashboard go to compute engine, now you're back at your overview click SSH Once connected to the server go on your screen where the folding is running to see the live action. be careful to exit with Ctrl+A+D NOT WITH ctrl+c!! With this configuration there wont be a log.txt. Dont fear its ok as long as you see it on "screen" command and under "top" command you see the usage everything is fine. How to change the configuration file (in case of error in passkey etc...) on the main prompt: Edit your config file with sudo nano /etc/fahclient/config.xml then there should be a line with <passkey v='xxxxxxxxx'/> (WITH both ' ) Correct it, save the file and restart the FAHClient to restart it go on the screen where it runs and kill it ctrl+c. then restart it on this same screen How to check CPU usage right at the server, in the main prompt top to geht out of there: crtl+c This is a very early and ruff version of this post and I am working on it to improve it! Anny questions, suggestions etc. Please let me know Happy folding, Metallus Note: Stepps 15 to the end and the tips sections should basically work on every Debian/u until etc. server How to do a quick PPD check Main Terminal: FAHClient --send-command <command> Commands: ppd -> gives you the current overall produced PPD configured -> if output is "True" your client is configured with username team etc slot-info -> gives you information on wich devices its running queue-info -> verry detailed info about the wokrunits being done at the moment help -> everything you can do and some you probably shouldn’t do. Here is proof of the first successful work unit calculated on my GCloud: