mersenneforum.org  

Go Back   mersenneforum.org > Search Forums

Showing results 1 to 25 of 39
Search took 0.01 seconds.
Search: Posts Made By: frmky
Forum: Msieve 2022-07-16, 22:58
Replies: 117
Views: 19,748
Posted By frmky
There's a good chance that won't work. The...

There's a good chance that won't work. The vectors are always kept on the card and may take most of the GPU memory, leaving little for the matrix blocks and spmv scratch space. Nothing beats...
Forum: Msieve 2022-05-10, 18:13
Replies: 117
Views: 19,748
Posted By frmky
That's a BIOS issue. Google says to look for...

That's a BIOS issue. Google says to look for options deep in the BIOS menus like PCI Express 64-bit BAR Support, large BARs, or above 4G decoding.
Forum: Msieve 2022-05-09, 22:08
Replies: 117
Views: 19,748
Posted By frmky
If nvidia-smi doesn't see it, then msieve won't....

If nvidia-smi doesn't see it, then msieve won't. Perhaps you need to reinstall the CUDA driver with the M40 installed?
Forum: Msieve 2022-05-01, 17:33
Replies: 117
Views: 19,748
Posted By frmky
GPU LA uses only a single CPU core to do a very...

GPU LA uses only a single CPU core to do a very small part of each iteration. Likewise, filtering and traditional sqrt use only a single core. The Core2 Duo should be fine.

With a 24GB card, you...
Forum: Msieve 2022-04-09, 19:45
Replies: 117
Views: 19,748
Posted By frmky
I sieve enough to use target_density of at least...

I sieve enough to use target_density of at least 100-110 as it brings down the matrix size. An 11 GB card can likely handle matrices with about 10M rows (GNFS-175ish), whereas a 24GB card would take...
Forum: Msieve 2022-03-27, 17:36
Replies: 117
Views: 19,748
Posted By frmky
You almost had enough. It ran out while trying to...

You almost had enough. It ran out while trying to allocate working memory for the spmv library. Recompile with VBITS=128 and it should fit, even if it's not optimal. (Don't forget to copy the .ptx...
Forum: Msieve 2022-03-25, 23:46
Replies: 117
Views: 19,748
Posted By frmky
That looks like the linux OOM killer. Which would...

That looks like the linux OOM killer. Which would mean the it has run out of available system (not GPU) memory.
Forum: Msieve 2022-02-24, 06:17
Replies: 117
Views: 19,748
Posted By frmky
Not at all! I look forward to seeing how you have...

Not at all! I look forward to seeing how you have gotten it to work in Colab!
Forum: Msieve 2022-02-21, 02:38
Replies: 117
Views: 19,748
Posted By frmky
Yes, it will work on a K80. My updated version...

Yes, it will work on a K80. My updated version requires CC 3.5 or greater.

You don't need to transfer the large relations file. Do this:
1. Complete the filtering and build the matrix locally....
Forum: Msieve 2021-10-31, 18:58
Replies: 117
Views: 19,748
Posted By frmky
It's done. 5RGLguge

It's done.
5RGLguge
Forum: Msieve 2021-10-26, 18:06
Replies: 117
Views: 19,748
Posted By frmky
For 2,2174L we sieved from 20M - 6B, and...

For 2,2174L we sieved from 20M - 6B, and collected 1.36B relations. This gave 734M uniques, so about 46% duplicates.

For 2,2174M we sieved from 20M - 4B, and collected 2.19B relations. This gave...
Forum: Msieve 2021-10-26, 07:48
Replies: 117
Views: 19,748
Posted By frmky
We didn't sieve it twice. Only a little at the...

We didn't sieve it twice. Only a little at the beginning was sieved with 33 bit LPs and all the relations were combined. There are a few stragglers that I'm not worrying about.
Forum: Msieve 2021-10-26, 04:05
Replies: 117
Views: 19,748
Posted By frmky
2,2174M is in LA, so here's one more data point....

2,2174M is in LA, so here's one more data point. Running on eight NVLink-connected V100's,
Sun Oct 24 01:15:27 2021 matrix is 106764994 x 106765194 (56998.7 MB) with weight 16127184931 (151.05/col)...
Forum: Msieve 2021-10-22, 23:00
Replies: 117
Views: 19,748
Posted By frmky
Technically it's an MxN matrix with M slightly...

Technically it's an MxN matrix with M slightly less than N, but for this question we can approximate it as NxN.

Volta (and I'm hoping Turing and Ampere) GPUs aren't very sensitive to the block_nnz...
Forum: Msieve 2021-09-24, 15:13
Replies: 117
Views: 19,748
Posted By frmky
A large fraction encounter issues when exceeding...

A large fraction encounter issues when exceeding 1GB/thread, so I stay a little below that.
Forum: Msieve 2021-09-24, 06:21
Replies: 117
Views: 19,748
Posted By frmky
And it's done. LA on the 102M matrix with...

And it's done. LA on the 102M matrix with restarts took 5 days 14 hours.
cB1qD1hJ
Forum: Msieve 2021-09-24, 06:17
Replies: 117
Views: 19,748
Posted By frmky
I'll try that, thanks!

I'll try that, thanks!
Forum: Msieve 2021-09-18, 15:59
Replies: 117
Views: 19,748
Posted By frmky
Does the lasieve5 code work correctly with 34-bit...

Does the lasieve5 code work correctly with 34-bit large primes? I know the check is commented out, but I haven't tested it.
Forum: Msieve 2021-09-18, 03:58
Replies: 117
Views: 19,748
Posted By frmky
For 2,2174L, 1355M relations yielded 734M...

For 2,2174L, 1355M relations yielded 734M uniques. With nearly 50% duplicates, we have clearly reached the limit for 16e. Anyway, filtering yielded
matrix is 102063424 x 102063602 (51045.3 MB) with...
Forum: Msieve 2021-09-17, 19:47
Replies: 117
Views: 19,748
Posted By frmky
No. The server is currently using 467G of 3.6T.

No. The server is currently using 467G of 3.6T.
Forum: Msieve 2021-09-17, 14:56
Replies: 117
Views: 19,748
Posted By frmky
CUDA doesn't allow display updates while a kernel...

CUDA doesn't allow display updates while a kernel is running. The only way to improve responsiveness without using a second GPU is to shorten the kernel run times. The longest kernel is the SpMV, and...
Forum: Msieve 2021-09-16, 06:00
Replies: 117
Views: 19,748
Posted By frmky
Today I expanded the allowed values of VBITS to...

Today I expanded the allowed values of VBITS to any of 64, 128, 192, 256, 320, 384, 448, or 512. This works on both CPUs and GPUs, but I don't expect much, if any, speedup on CPUs. As a GPU...
Forum: Msieve 2021-09-13, 05:22
Replies: 117
Views: 19,748
Posted By frmky
I spent time with Nsight Compute looking at the...

I spent time with Nsight Compute looking at the SpMV kernel. As expected for SpMV it's memory bandwidth limited, so increasing occupancy to hide latency should help. I adjusted parameters to reduce...
Forum: Msieve 2021-08-11, 22:09
Replies: 117
Views: 19,748
Posted By frmky
The LA for 2,2162M, an 84.2M matrix, successfully...

The LA for 2,2162M, an 84.2M matrix, successfully completed on four NVLink-connected V100's in a total of 95.5 hours of runtime. There was a restart due to the 48-hour queue time limit on SDSC...
Forum: Msieve 2021-08-08, 00:57
Replies: 117
Views: 19,748
Posted By frmky
Yes, you could solve a matrix up to about 15M or...

Yes, you could solve a matrix up to about 15M or so on the card. If you have at least 32 GB system memory, you could go a bit larger transferring the matrix from system memory as needed using CUDA...
Showing results 1 to 25 of 39

 
All times are UTC. The time now is 20:36.


Mon Dec 5 20:36:04 UTC 2022 up 109 days, 18:04, 0 users, load averages: 1.24, 1.05, 0.93

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.

≠ ± ∓ ÷ × · − √ ‰ ⊗ ⊕ ⊖ ⊘ ⊙ ≤ ≥ ≦ ≧ ≨ ≩ ≺ ≻ ≼ ≽ ⊏ ⊐ ⊑ ⊒ ² ³ °
∠ ∟ ° ≅ ~ ‖ ⟂ ⫛
≡ ≜ ≈ ∝ ∞ ≪ ≫ ⌊⌋ ⌈⌉ ∘ ∏ ∐ ∑ ∧ ∨ ∩ ∪ ⨀ ⊕ ⊗ 𝖕 𝖖 𝖗 ⊲ ⊳
∅ ∖ ∁ ↦ ↣ ∩ ∪ ⊆ ⊂ ⊄ ⊊ ⊇ ⊃ ⊅ ⊋ ⊖ ∈ ∉ ∋ ∌ ℕ ℤ ℚ ℝ ℂ ℵ ℶ ℷ ℸ 𝓟
¬ ∨ ∧ ⊕ → ← ⇒ ⇐ ⇔ ∀ ∃ ∄ ∴ ∵ ⊤ ⊥ ⊢ ⊨ ⫤ ⊣ … ⋯ ⋮ ⋰ ⋱
∫ ∬ ∭ ∮ ∯ ∰ ∇ ∆ δ ∂ ℱ ℒ ℓ
𝛢𝛼 𝛣𝛽 𝛤𝛾 𝛥𝛿 𝛦𝜀𝜖 𝛧𝜁 𝛨𝜂 𝛩𝜃𝜗 𝛪𝜄 𝛫𝜅 𝛬𝜆 𝛭𝜇 𝛮𝜈 𝛯𝜉 𝛰𝜊 𝛱𝜋 𝛲𝜌 𝛴𝜎𝜍 𝛵𝜏 𝛶𝜐 𝛷𝜙𝜑 𝛸𝜒 𝛹𝜓 𝛺𝜔