-
-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
-
mixbench
A GPU benchmark tool for evaluating GPUs and CPUs on mixed operational intensity kernels (CUDA, OpenCL, HIP, SYCL, OpenMP)
And how would you go implementing this on a GPU? I could also dig up some SIMD accelerated entropy coding. Impossible to run performantly on a GPU regardless of the amount of data. Video encoding/decoding also don't run well on GPUs, thus the dedicated hardware blocks. If you want to software decode AV1 for example, AVX512 helps a lot, and you can't code it in CUDA performantly, for example.
If you're looking for an example, perhaps the despacer problem might be one which doesn't get too complex. Do you know of a way to implement it on a GPU such that it'd run better than (or at least as good as) a CPU SIMD implementation would?
The results I get match the FLOPS figures stated for the respective GPUs, so presumably I can't be memory bound or similar. But if you're still in doubt, I was using this code, comparing the single precision and integer kernels, so let me know any issues you see with the benchmark.