2 ms·
The cuDF interop in the roadmap [1] will be huge for my workloads. XGBoost has the fastest inference time on GPUs, so a fast path straight from these Vortex fil
by kipukun 11mo ago
The cuDF interop in the roadmap [1] will be huge for my workloads. XGBoost has the fastest inference time on GPUs, so a fast path straight from these Vortex files to GPU memory seems promising.
[1] https://github.com/vortex-data/vortex/issues/2116 https://github.com/vortex-data/vortex/issues/2116
- reactordev 11mo agoCan you explain how it’s faster? GPU memory is just a blob with an address. Is it because the loading algorithms for vortex align better with XGBoost or just plain uploading to the GPU?
- robert3005 11mo agoWhat you can do if you have gpu friendly format is you send compressed data over PCI-E and then decompress on the gpu. Thus your overall throughput will increase since PCI-E bandwidth is the limiting factor of the overall system.
- reactordev 11mo agoThat doesn’t explain how vortex is faster. Yes, you should send compressed data to the GPU and let it uncompress. You should maximize your PCI-E throughput to minimize latency in execution, but what does Vortex bring? Other than Parque bad, Vortex good.
- kipukun 11mo agoXGBoost is just faster on the GPU, regardless of the file format. A sibling post also pointed out compression helping out on bandwidth.