Post
Topic
Board Development & Technical Discussion
Merits 3 from 1 user
Re: Pollard's kangaroo ECDLP solver
by
MrFreeDragon
on 04/05/2020, 12:56:55 UTC
⭐ Merited by ETFbitcoin (3)
This is a benchmark post with speeds for Kangaroo GPU Solver. All the tests were made with default DP and default grid size (calculated by a program). I guess that some plays with DP and grid size could change (increase or decrease the speed). Is somebody knows the optimal values, please let us know.

Code:
  Card Model          Grid size      DP        Tested speed
---------------------------------------------------------------
GTX 1050 Ti          Grid(12x256)   DP 16     115 MKey/sec
GTX 1080 Ti          Grid(56x256)   DP 15     500 MKey/sec
Tesla T4 16Gb        Grid(80x128)   DP 14     565 MKey/sec
RTX 2080ti 11Gb      Grid(136x128)  DP 13     1225 MKey/sec
Tesla V100 32Gb      Grid(160x128)  DP 13     1420 MKey/sec
---------------------------------------------------------------

It looks like the Kangaroo solver works fine with GeForce family cards but not with Quadro/NVS family. For example Tesla V100 is 4-5 times more expensive than RTX 2080ti, but the speed is only 15% higher.

And we also can learn from this table that the most efficient card for kangaroo is RTX 2080ti 11Gb which has the best performance in terms of speed/cost.