This Week in MLC@Home
Notes for Nov 30 2020
A weekly summary of news and notes for MLC@Home

Summary
Only a short update this week, as we've been running analysis on the data collected so far in preparation for a paper, and it was a holiday weekend in our part of the world. The new Linux/CUDA client appears to be working well, so we can stop spamming the news feed with updated each week. More information in the news below.

News:

• More details on the updated Linux GPU support, both CUDA and ROCm, and it is here, including a recent discussion on disk usage of the new client: https://www.mlcathome.org/mlcathome/forum_thread.php?id=127.
  
• DS1 and DS2 continue towards completion. We've modified some of the WUs to expire early if they don't appear to be converging, in the hopes of nudging our searches down more likely-to-converge-soon paths. EightBitModified is only a few thousand away from completion!
  
• DS3 progress is approaching milestone 2, 1000x100.. 100000 DS3 networks trained. That will definitely help with our ongoing analysis. It's nice to see the scoreboard at the bottom of https://www.mlcathome.org/ turning from yellow to green!
  
• Analyzing the DS3 data so far, we've discovered the dataset itself is very large (approx 100GB for just 500x100). This is slowing down analysis somewhat, but a good problem to have. We can trim that down significantly in a release, but sometimes dealing with that much data just takes time.
  
• We've created the DS1-100, DS1-500, DS1-1000, DS2-100, DS3-100, DS3-500 datasets internally, and are cleaning them up and analyzing the results. We hope to have some (preliminary) results to share with the community, like we published before, within days.
  

Project status snapshot:
(note these numbers are approximations)

Tasks
Tasks ready to send 45843
Tasks in progress 18569
Users
With credit 1318
Registered in past 24 hours 43/b]
[b]Hosts
With recent credit 2228
Registered in past 24 hours 60
Current GigaFLOPS 37671.35

Dataset 1 and 2 progress:

SingleDirectMachine      10002/10004
EightBitMachine          10001/10006
SingleInvertMachine      10001/10003
SimpleXORMachine         10000/10002
ParityMachine             1091/10005

ParityModified             391/10005
EightBitModified          7750/10006
SimpleXORModified        10005/10005
SingleDirectModified     10004/10004
SingleInvertModified     10002/10002 

Dataset 3 progress:

Overall (so far): 82873/101488
Milestone 1, 100x100:  10000/10000
Milestone 2, 100x1000: 82873/100000
Milestone 3: 100x10000: 82873/1000000

Last week's TWIM Notes: Nov 23 2020

Thanks again to all our volunteers!

-- The MLC@Home Admins(s)
Homepage: https://www.mlcathome.org/
Twitter: @MLCHome2

Nice to hear that the Linux/CUDA client is working fine. Thanks!

---

Thanks for the update. Even though the snapshot data you include is at best indicative, the recent trend in the compute power development seems promising. Looks like the GPU rollout really starts to speed things up.

Current GigaFLOPS
10/12: 24544.93
10/19: 248734.27
10/27: 30532.46
11/2: 29333.56
11/9: 26267.99
11/17: 33798.72
11/23: 36126.42
11/30: 37671.35

This week's combined compute power is ~53% higher than only 1.5 months ago. While Mumps is still pulling away with a comfortable lead, others are picking up. It seems that Mumps will be the first to reach the 500m badge. Interesting though that this project's top daily credits days all date back to the beginning of the project...https://www.boincstats.com/stats/190/project/detail/bestxdays
I suspected that GPUs would do the trick and quickly take the lead.