lehuyduc
lehuyduc
@gbanjac hello, I just run OSQP GPU branch `cuda-1.0` on my machine, and it's actually 2x slower on 4000x4000 input matrix. When I run `nvidia-smi`, I also found that GPU...
> My biggest question there is around administering the thing (e.g. how to disable turbo-boost and SMT, which would be a good idea), as I'm not super-savvy when it comes...
Could you share your computer spec please? Thanks!
@AlexanderYastrebov @gunnarmorling could you check please? Thanks!
Please don't take it the wrong way 🥲 I really want to see what creative solutions you come up for hash map, that's why I set it to 500M exactly...
I can do that but the goal isn't to kill creativity 🥳 So I will stick to 500M exactly (please don't read the file in reverse 😟)
Ran your latest code again! On Threadripper 5995WX this time. Result at the end ``` #include #include #include #include #include #include #include #include #include #include #include #define MAX_DISTINCT_GROUPS 10000 #define...
> Didn't except such a big improvement from those last few micro-optimizations. Same. A few days ago I thought I've squeezed out everything, but slowly my final time went from...
I think all of us need to share our datasets :rofl: Both `./create_measurements.sh` and `./create_measurements3.sh` are non-deterministic, so everyone has their own datasets and optimize our hash tables for the...