| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: OpenPandemics - COVID-19 Project
Thread: One terabyte written to my SSD since the stress test began |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 81
|
|
| Author |
|
|
goben_2003
Advanced Cruncher Joined: Jun 16, 2006 Post Count: 146 Status: Offline Project Badges: 
|
I have a problem after using Ramdisk, I'm only getting 1 or 2 WUs with each work request, I'm slowly finishing more than I'm getting back, even though I only have 20 WUs in cache, I should have up to 150 in cache. My Ramdisk is 4GB and there's more than 1GB free and my settings allow to use it all. This is from the work_fetch_debug: Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | checking NVIDIA GPU Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | [work_fetch] set_request() for NVIDIA GPU: ninst 3 nused_total 9.00 nidle_now 0.00 fetch share 1.00 req_inst 0.00 req_secs 55729.12 Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | NVIDIA GPU set_request: 55729.123704 Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | [work_fetch] request: CPU (0.00 sec, 0.00 inst) NVIDIA GPU (55729.12 sec, 0.00 inst) Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | Sending scheduler request: To fetch work. Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | Reporting 4 completed tasks Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | Requesting new tasks for NVIDIA GPU Fri 30 Apr 2021 21:20:25 CEST | World Community Grid | Scheduler request completed: got 2 new tasks Fri 30 Apr 2021 21:20:25 CEST | | [work_fetch] Request work fetch: RPC complete Any idea what's wrong? It looks like WCG just is not sending as much work as you requested. I think it has nothing to do with the ramdisk. If it was not enough free space, there would be a message in the log. Something along the lines of OPNG requires X space free, but there is only Y.  |
||
|
|
goben_2003
Advanced Cruncher Joined: Jun 16, 2006 Post Count: 146 Status: Offline Project Badges:
|
Reguarding ImDisk: "Dynamic ramdisks do not work properly when they are mounted on a folder instead of a drive letter, and if you don't use the TRIM commands (TRIM commands are used by default with NTFS on Windows 7 and later)." -- https://sourceforge.net/p/imdisk-toolkit/doc/Home/ Ok, I guess then there are 2 questions: 1. Do Andyman or TripleTrouble have TRIM disabled? It should be enabled for Andyman unless he disabled it since he is running Windows 10 pro. It can be checked by opening an Administrator command prompt and running: fsutil behavior query DisableDeleteNotify If DisableDeleteNotify=0, then TRIM is enabled. 2. Do dynamic ramdisks sometimes have issues even if TRIM is enabled? I am not running it as a dynamic ramdisk. Edit: minor correction [Edit 1 times, last edit by goben_2003 at Apr 30, 2021 7:55:13 PM] |
||
|
|
Etul
Cruncher Australia Joined: Sep 12, 2014 Post Count: 38 Status: Offline Project Badges: 
|
Just wondering if it would be worth making a separate post with the Imdisk instructions, I feel like a lot of people might miss this really useful information when it is stuck on page 2. :)
---------------------------------------- |
||
|
|
JohnDK
Advanced Cruncher Denmark Joined: Feb 17, 2010 Post Count: 77 Status: Offline Project Badges:
|
I have a problem after using Ramdisk, I'm only getting 1 or 2 WUs with each work request, I'm slowly finishing more than I'm getting back, even though I only have 20 WUs in cache, I should have up to 150 in cache. My Ramdisk is 4GB and there's more than 1GB free and my settings allow to use it all. This is from the work_fetch_debug: Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | checking NVIDIA GPU Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | [work_fetch] set_request() for NVIDIA GPU: ninst 3 nused_total 9.00 nidle_now 0.00 fetch share 1.00 req_inst 0.00 req_secs 55729.12 Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | NVIDIA GPU set_request: 55729.123704 Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | [work_fetch] request: CPU (0.00 sec, 0.00 inst) NVIDIA GPU (55729.12 sec, 0.00 inst) Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | Sending scheduler request: To fetch work. Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | Reporting 4 completed tasks Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | Requesting new tasks for NVIDIA GPU Fri 30 Apr 2021 21:20:25 CEST | World Community Grid | Scheduler request completed: got 2 new tasks Fri 30 Apr 2021 21:20:25 CEST | | [work_fetch] Request work fetch: RPC complete Any idea what's wrong? It looks like WCG just is not sending as much work as you requested. I think it has nothing to do with the ramdisk. If it was not enough free space, there would be a message in the log. Something along the lines of OPNG requires X space free, but there is only Y. It's just strange, there was no problem with work fetch before and the only setting I've change after using Ramdisk, was to allow to use all disk space. Have changed back to the HD becourse I was down to only 4 WUs, and the first work request gave 147 new WUs. Anybody able to explain that??
Intel i7-6850K / 16GB / RTX 3090 / 2x RTX 3080 Ti / RTX 3070 Ti
AMD Ryzen 9 5950X / 32GB / RTX 2080 Ti |
||
|
|
goben_2003
Advanced Cruncher Joined: Jun 16, 2006 Post Count: 146 Status: Offline Project Badges:
|
Just wondering if it would be worth making a separate post with the Imdisk instructions, I feel like a lot of people might miss this really useful information when it is stuck on page 2. :) Possibly, but I think we should work through issues that people are having first to make sure the instructions are correct. |
||
|
|
goben_2003
Advanced Cruncher Joined: Jun 16, 2006 Post Count: 146 Status: Offline Project Badges:
|
I have a problem after using Ramdisk, I'm only getting 1 or 2 WUs with each work request, I'm slowly finishing more than I'm getting back, even though I only have 20 WUs in cache, I should have up to 150 in cache. My Ramdisk is 4GB and there's more than 1GB free and my settings allow to use it all. This is from the work_fetch_debug: Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | checking NVIDIA GPU Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | [work_fetch] set_request() for NVIDIA GPU: ninst 3 nused_total 9.00 nidle_now 0.00 fetch share 1.00 req_inst 0.00 req_secs 55729.12 Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | NVIDIA GPU set_request: 55729.123704 Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | [work_fetch] request: CPU (0.00 sec, 0.00 inst) NVIDIA GPU (55729.12 sec, 0.00 inst) Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | Sending scheduler request: To fetch work. Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | Reporting 4 completed tasks Fri 30 Apr 2021 21:20:23 CEST | World Community Grid | Requesting new tasks for NVIDIA GPU Fri 30 Apr 2021 21:20:25 CEST | World Community Grid | Scheduler request completed: got 2 new tasks Fri 30 Apr 2021 21:20:25 CEST | | [work_fetch] Request work fetch: RPC complete Any idea what's wrong? It looks like WCG just is not sending as much work as you requested. I think it has nothing to do with the ramdisk. If it was not enough free space, there would be a message in the log. Something along the lines of OPNG requires X space free, but there is only Y. It's just strange, there was no problem with work fetch before and the only setting I've change after using Ramdisk, was to allow to use all disk space. Have changed back to the HD becourse I was down to only 4 WUs, and the first work request gave 147 new WUs. Anybody able to explain that?? Did you have "Allocate Memory Dynamically" checked? I know there was a note that having that checked makes the ramdisk slower. Not sure that it makes a difference though. I did not have that checked and I have stayed at the max 50 WUs / GPU. I am running 10 at a time on my nvidia rtx 2060m. |
||
|
|
JohnDK
Advanced Cruncher Denmark Joined: Feb 17, 2010 Post Count: 77 Status: Offline Project Badges:
|
Don't know where the "Allocate Memory Dynamically" setting is located.
----------------------------------------
Intel i7-6850K / 16GB / RTX 3090 / 2x RTX 3080 Ti / RTX 3070 Ti
AMD Ryzen 9 5950X / 32GB / RTX 2080 Ti |
||
|
|
goben_2003
Advanced Cruncher Joined: Jun 16, 2006 Post Count: 146 Status: Offline Project Badges:
|
Don't know where the "Allocate Memory Dynamically" setting is located. It is on the "Basic" tab of "Imdisk configuration" from step 4. |
||
|
|
MJH333
Senior Cruncher England Joined: Apr 3, 2021 Post Count: 268 Status: Offline Project Badges: 
|
Hi goben_2003
Thank you very much for the instructions on how to create a RAM disk. I have tried this out and it seems to be working. It took me a little while to realise that although I had set the RAM disk to 4GB, I had also set the Computing preferences in the BOINC Manager to "Use no more than 50% of total". So I had accidentally reduced the size of the RAM disk to 2GB! (Oops!). I wonder if it would be worth mentioning that trap in your instructions. Cheers!  |
||
|
|
goben_2003
Advanced Cruncher Joined: Jun 16, 2006 Post Count: 146 Status: Offline Project Badges:
|
Hi goben_2003 Thank you very much for the instructions on how to create a RAM disk. I have tried this out and it seems to be working. It took me a little while to realise that although I had set the RAM disk to 4GB, I had also set the Computing preferences in the BOINC Manager to "Use no more than 50% of total". So I had accidentally reduced the size of the RAM disk to 2GB! (Oops!). I wonder if it would be worth mentioning that trap in your instructions. Cheers! Hi MJH333, You're welcome! Thank you for pointing it out to help others! I added a note about it to the instructions. |
||
|
|
|