I have a dual xeon CPU machine with 64GB RAM which i'm about to double to 128GB. What memory intensive WUs can I volunteer for? Just jumped from single to dual processors but now have lots of RAM sitting idle. 128GB is almost as much as my original boot drive until i did a swap to my board 260 GB mSATA which is half as fast as the old boot drive. Doh! Board mSATA seems to 3GB/s max.
Never mind noise in the background for current Androids.
RAM heavy stuff sought before i get bored and try to run a multi nested Linux VM in a RAM drive.
RAM heavy points
Forum rules
Welcome to The Scottish Boinc Team boards. See forum rules in pinned post. If you can't be bothered then try not to be too naughty as I have a delete button to press and a ban hammer to swing.
Welcome to The Scottish Boinc Team boards. See forum rules in pinned post. If you can't be bothered then try not to be too naughty as I have a delete button to press and a ban hammer to swing.
-
- Boinc Recruit
- Posts: 2
- Joined: Mon Jun 01, 2020 1:35 pm
#1 RAM heavy points
Lenovo C30: 2x Xeon e5-2680v2 20 cores 40 virtual processors 64GB ECC RDIMM RAM NVIDIA GTX 1650 Super OC
- scole of TSBT
- Boinc Major General
- Posts: 5980
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#2 Re: RAM hevy points
Try the Yoyo ECM 2 WUs. They can use 10GB RAM per WU. Be sure to set your your boinc preferences to not use more Memory than what is needed to keep your server from swapping memory and becoming unresponsive though. Set Max Memory Usage to maybe 90% so boinc leaves enough to keep system running smooth.
-
- Boinc Recruit
- Posts: 2
- Joined: Mon Jun 01, 2020 1:35 pm
#3 Re: RAM hevy points
10 gigs per WU?
I've been running it but don't see this. My other project has dried up and its had 100% for yoyo but i see little increase in RAM demand in my monitoring software.
My planned upgrade to 128GB was a bit of a SNAFU. Misidentified the new ram as RDIMM when it was LRDIMM due to out of focus pic on ebay.
I've been running it but don't see this. My other project has dried up and its had 100% for yoyo but i see little increase in RAM demand in my monitoring software.
My planned upgrade to 128GB was a bit of a SNAFU. Misidentified the new ram as RDIMM when it was LRDIMM due to out of focus pic on ebay.
Lenovo C30: 2x Xeon e5-2680v2 20 cores 40 virtual processors 64GB ECC RDIMM RAM NVIDIA GTX 1650 Super OC
- scole of TSBT
- Boinc Major General
- Posts: 5980
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#4 Re: RAM heavy points
Are you sure you are running ECM 2 and not the other apps? Do you have a link to a result?
EDIT:
They don't appear to have many ECM 2 WUs currently.
NFS lasievef app uses around a 1 GB per WU. If you are comfortable running projects that use virtual box, LHC's Atlas app uses a lot of RAM but you will need to setup an app_config.xml else it will allocate a bunch of RAM but will only use a fraction of it. We can step you though it if you want to try it.
EDIT:
They don't appear to have many ECM 2 WUs currently.
NFS lasievef app uses around a 1 GB per WU. If you are comfortable running projects that use virtual box, LHC's Atlas app uses a lot of RAM but you will need to setup an app_config.xml else it will allocate a bunch of RAM but will only use a fraction of it. We can step you though it if you want to try it.
#5 Re: RAM heavy points
I just ran some NanoHub Wu and they were taking up to 4G /WU.