ECM for CUDA GPUs in latest GMP-ECM ?

[VBCurtis]

The GPU's advantage is running many curves in parallel; pm1 does a single curve, so a GPU would be a total waste for pm1.

[ET_]

Originally Posted by VBCurtis

The GPU's advantage is running many curves in parallel; pm1 does a single curve, so a GPU would be a total waste for pm1.

...unless you use CUDA-Pm1 that parallelizes the FFT calculations...
But that's a totally different subject.

[pepi37]

Originally Posted by VBCurtis

The GPU's advantage is running many curves in parallel; pm1 does a single curve, so a GPU would be a total waste for pm1.

Many curves in parallel until limit.

[pepi37]

Originally Posted by ET_

...unless you use CUDA-Pm1 that parallelizes the FFT calculations...
But that's a totally different subject.

Some more info? Link please!

That program is only for base2, I am doing outside that :)

[xilman]

Originally Posted by pepi37

So GPU has limit, when you use ECM, and what is about -pm1?
In that case is same limit ?

The limit is on the arithmetic, not the algorithms built on top of it.

[pepi37]

Originally Posted by xilman

The limit is on the arithmetic, not the algorithms built on top of it.

So for now GPU as option in GMP-EMC is no choice for me :(
Continuing to explore GMP-EMC.....

[bsquared]

Maybe I missed it, but I haven't seen any benchmark numbers for GPU-ECM since several years ago in this thread. Could I ask someone to do that for current-generation GPUs? e.g., how long does it take to run N parallel curves for each of the 512-bit and 1018-bit versions (which I understand to be the only reliable ones)?

[wombatman]

Originally Posted by bsquared

Maybe I missed it, but I haven't seen any benchmark numbers for GPU-ECM since several years ago in this thread. Could I ask someone to do that for current-generation GPUs? e.g., how long does it take to run N parallel curves for each of the 512-bit and 1018-bit versions (which I understand to be the only reliable ones)?

Got a particular number (or size of number) in mind?

[bsquared]

Originally Posted by wombatman

Got a particular number (or size of number) in mind?

The way I understand things, it doesn't matter; any number will run in the same amount of time (although different for the different versions 512 vs. 1018-bit). Someone let me know if that's not the case.

I'm currently working on the 197 digit cofactor of 149^70+70^149:

18990123508557902868419834986612849212629329047848408031918356871905091180915018818857084783656883790655928770567349115604303665109625610225945717466120031386943078873289069222693826896799892056101

which you can use for the larger ECM version. Pick anything you like for the smaller one.

[wombatman]

Originally Posted by bsquared

The way I understand things, it doesn't matter; any number will run in the same amount of time (although different for the different versions 512 vs. 1018-bit). Someone let me know if that's not the case.

I'm currently working on the 197 digit cofactor of 149^70+70^149:

18990123508557902868419834986612849212629329047848408031918356871905091180915018818857084783656883790655928770567349115604303665109625610225945717466120031386943078873289069222693826896799892056101

which you can use for the larger ECM version. Pick anything you like for the smaller one.

I'll run the lower levels tonight as I get a chance to. That will provide an approximate trend, as I've noticed that the time involved for higher B1s is roughly linear with B1.

[xilman]

Originally Posted by bsquared

Maybe I missed it, but I haven't seen any benchmark numbers for GPU-ECM since several years ago in this thread. Could I ask someone to do that for current-generation GPUs? e.g., how long does it take to run N parallel curves for each of the 512-bit and 1018-bit versions (which I understand to be the only reliable ones)?

I'll see what I can do. For better or worse I seem to have ended up with the job of maintaining GPU-ECM. Any assistance with that task will be much appreciated. In particular, evidence of inadequacies of the software will be useful. Even more useful will be contributions to its enhancement.

BTW, and AFAIK, the 512-bit version is really 506-bit limited. One of the things on my to-do list is to allow for more versions. Ideally the end-user shouldn't have to predetermine the size of the arithmetic. Another WIBNI is to implement stage 2 on the GPU. That might be the easier of the two. I would also like to extend the ECMNET client to use a GPU where available.