-
**Is your feature request related to a problem? Please describe.**
i have a myriad of software servers on my home network, i was pressured into sorting out SSL only to find an absolute mess of certif…
-
The daily score calculation scales exponentially with the number of models and benchmarks, and is likely the cause of the daily "slow loading" we are currently seeing (previously resulting in regular …
-
The Nvidia HPCG benchmark used to available only via containers, but it was open-sourced last week and it's available at https://github.com/NVIDIA/nvidia-hpcg. This works on both CPU (x86-64 and aarc…
-
Currently the document processing pipeline has an issue, where sometimes in the process of writing a script there will be some obvious error with the documents.
Currently the solution has been to …
-
I just noticied that the JSC easyconfigs set a number of additional options for Clang 9.0.0:
```
configopts = '-DCLANG_OPENMP_NVPTX_DEFAULT_ARCH=sm_37'
configopts += ' -DLIBOMPTARGET_NVPTX_COMPUTE_…
-
Building an OpenMP hello world fails to link, when Optimization is turned on:
```c
#include
#include
int main(int argc, char** argv){
printf("Devices: %i\n", omp_get_num_devices());
int a[…
-
### Description
Approximately 15% performance regression in transaction handling is caused by enabling `streamMode = BOTH` in #16167.
The wall clock time cost is attributed to `Hedera.manageBlockEnd…
-
### Background and motivation
`memmove` and `memset` are the most common operations from the earliest days of computing to today.
They are included in most practical programming languages, except as…
-
```
➜ go-ggllm.cpp git:(master) ✗ CGO_LDFLAGS="-lcublas -lcudart -L/usr/local/cuda/lib64/" LIBRARY_PATH=$PWD C_INCLUDE_PATH=$PWD go run ./examples -m "/media/NAS/MLModels/02_LLMs/falcon-40b-instruct…
-
Existing approach is that if users specify `offloadNone()` execution strategy on the builder, we respect that and disable offloading. However, if they have a filter that for sure has blocking code, it…