Inference Llama 2 in one file of pure C

NVIDIA Linux open GPU kernel module source

Checkpoint/Restore tool

Envision a world where EVERY student can read ALL the code of a teaching operating system.

RDMA core userspace libraries and daemons

Unified Communication X (mailing list - https://elist.ornl.gov/mailman/listinfo/ucx-group)

User space software for Intel(R) Resource Director Technology

Mirror of the official Barrelfish OS repository.

RDMA and SHARP plugins for nccl library

Interference-aware CPU scheduling that enables performance isolation and high CPU utilization for datacenter servers

example code for using DC QP for providing RDMA READ and WRITE operations to remote GPU memory

Arbitrary offloads for RDMA NICs

An attempt at achieving the theoretical best memory bandwidth of my machine.

A GPU-accelerated DNN inference serving system that supports instant kernel preemption and biased concurrent execution in GPU scheduling.

A GPU-Accelerated In-Memory Key-Value Store (AWS-focused fork)