Evaluate and test scheduling algorithms for ERDOS

Official release of DepS: Delayed Eps-Shrinking for Faster Once-For-All Training, ECCV 2024

LLM Serving Performance Evaluation Harness

High-speed Large Language Model Serving for Local Deployment

flame is a federated learning system for edge with flexibility and scalability at the core of its design.

[ICLR 2021] CompOFA: Compound Once-For-All Networks For Faster Multi-Platform Deployment

A low-latency & high-throughput serving engine for LLMs

A large-scale simulation framework for LLM inference

ACM SoCC Top-50 authors

Modin: Scale your Pandas workflows by changing a single line of code

Serverless ML Framework

Deadline-based hyperparameter tuning on RayTune.

Prediction Serving on Ray

cloc counts blank lines, comment lines, and physical lines of source code in many programming languages.

A low-latency prediction-serving system

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Google's Operations Research tools:

Mass Parallel SSH