🏠
Highlights
Lists (1)
Starred repositories
An eBPF based tool for implementing socket based communication to shared memory
Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.
Inspect a command's effects before modifying your live system
Userspace eBPF instrumentation and control framework for deploying control and monitoring functions in a secure manner
Dynamic Memory Management for Serving LLMs without PagedAttention
GPU-accelerated vector query processing system that supports large vector datasets beyond GPU memory.
eBPF for GPU UVM offloading and scheduling in Linux kernel
System Intelligence Benchmark
A library for efficient similarity search and clustering of dense vectors.
Paella: Low-latency Model Serving with Virtualized GPU Scheduling
cache_ext is a framework to customize Linux page cache eviction policies using BPF. Appeared in SOSP 2025.
DISB is a new DNN inference serving benchmark with diverse workloads and models, as well as real-world traces.
A fast GPU memory copy library based on NVIDIA GPUDirect RDMA technology
A language and library for specifying syscall filtering policies.
cvc5 is an open-source automatic theorem prover for Satisfiability Modulo Theories (SMT) problems.
FlashInfer: Kernel Library for LLM Serving
Python SDK for ACP clients and agents.
This is the public repo for DREAM accepted to ICS 2025.
Cost-efficient Out-of-core GNN Training System on TB-scale Graph [ICDE 25]
Mirage Persistent Kernel: Compiling LLMs into a MegaKernel
