HN Explorer

medicis123

↗ HN profile [ 36.4 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. 1 points 0
  2. 1 points 0
  3. Show HN: Stop GPU pods placement getting bottlenecked by reserved VRAM

    We have built a GPU Runtime for Nvidia GPUs that can run multiple development/experimental/inference workloads per GPU with safe overcommit of VRAM, dynamic fractional allocation of GPU cores, and Deduplication of…

    2 points 0
  4. A New Approach to GPU Sharing: Deterministic, SLA-Based GPU Kernel Scheduling

    Most GPU “sharing” solutions today (MIG, time-slicing, vGPU, etc.) still behave like partitions: you split the GPU or rotate workloads. That helps a bit, but it still leaves huge portions of the GPU idle and introduces…

    1 points 0
  5. Show HN: Disaggregating GPU compute from CPU in ML job execution to scale GPUs (woolyai.com)

    WoolyAI disaggregates GPU compute from CPU and routes GPU operations from ML jobs running on CPU-only infra into a shared heterogeneous GPU pool(Nvidia+AMD), where a GPU hypervisor packs and schedules multiple jobs per…

    1 points 0
  6. Show HN: Run PyTorch on CPU boxes, offload kernels to remote GPUs

    We have opened the WoolyAI GPU hypervisor trial to all. https://woolyai.com/signup/ - Higher GPU utilization & lower cost Pack many jobs per GPU with WoolyAI’s server-side scheduler, VRAM deduplication, and SLO-aware…

    1 points 0
  7. Running Nvidia CUDA PyTorch container project/pipelines on AMD with no changes

    Hi, I wanted to share some information on this cool feature we built in WoolyAI GPU hypervisor, which enables users to run their existing Nvidia CUDA pytorch/vLLM projects and pipelines without any modifications on AMD…

    1 points 0
  8. GPU-accelerated code on CPU-only environments -Remote GPU Kernel Execution (youtube.com)
    1 points 1
  9. Sharing base model in GPU VRAM across multiple inference stack process [video] (youtube.com)
    7 points 1
  10. Sharing actual GPU core and VRAM utilization metrics for query on 10 LLM models (woolyai.com)
    1 points 2
  11. Show HN: WoolyAI-CUDA Abstraction Layer to Decouple Kernel Shader Exec on GPU (woolyai.com)

    Hi HN, We’re a small team of OS, virtualization, and ML engineers, and after three years of development, we’re thrilled to launch the beta of our CUDA abstraction layer! We decouple the Kernel Shader execution from…

    4 points 0
  12. Locally delivered and centrally managed macOS envs for privileged access setup (veertu.com)
    1 points 0
  13. Shopify scaling iOS CI with Anka (engineering.shopify.com)
    1 points 0