Submissions by mlerner

Autothrottle: Resource Management for SLO-Targeted Microservices

Resiliency at Scale: Managing Google's TPUv4 Machine Learning Supercomputer

How the anti-cheat is anti-cheating so far

ServiceRouter: Hyperscale and Minimal Cost Service Mesh at Meta

A Cloud-Scale Characterization of Remote Procedure Calls

A Cloud-Scale Characterization of Google's Remote Procedure Calls

Gemini, Amazon's system for fast failure recovery in distributed model training

Defcon: Preventing overload with graceful feature degradation (2023)

MotherDuck: DuckDB in the Cloud and in the Client [pdf]

Gemini: Fast Failure Recovery in Distributed Training with In-Memory Checkpoints

Gemini: Fast Failure Recovery in Distributed Training with In-Memory Checkpoints

XFaaS: Hyperscale and Low Cost Serverless Functions at Meta

XFaaS: Hyperscale and Low Cost Serverless Functions at Meta

Blueprint: A Toolchain for Highly-Reconfigurable Microservice Applications

Gemini: Fast Failure Recovery in Distributed Training with In-Memory Checkpoints [pdf]

XFaaS: Hyperscale and Low Cost Serverless Functions at Meta [pdf]

Efficient Memory Management for Large Language Model Serving with PagedAttention

Efficient Memory Management for Large Language Model Serving with PagedAttention

Blueprint: A Toolchain for Highly-Reconfigurable Microservice Applications

Blueprint: A Toolchain for Highly-Reconfigurable Microservice Applications

Defcon: Preventing Overload with Graceful Feature Degradation

Defcon: Preventing Overload with Graceful Feature Degradation