pentium3

icon indicating a website link https://pentium3.github.io/ icon indicating an email [email protected]

icon location Boston icon location emmmmmmmmmm

Results 104 issues of pentium3

PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel

https://arxiv.org/pdf/2304.11277.pdf

Splitwise: Efficient Generative LLM Inference Using Phase Splitting

https://arxiv.org/pdf/2311.18677.pdf

Computing in the Era of Large Generative Models: From Cloud-Native to AI-Native

https://arxiv.org/pdf/2401.12230.pdf

FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

https://github.com/Dao-AILab/flash-attention

llm

Efficient Memory Management for Large Language Model Serving with PagedAttention

1 comment

https://arxiv.org/pdf/2309.06180.pdf

llm
SysforML
sosp23

tf.data service: A Case for Disaggregating ML Input Data Processing

https://arxiv.org/pdf/2210.14826.pdf

socc23

Gödel: Unified Large-Scale Resource Management and Scheduling at ByteDance

https://dl.acm.org/doi/pdf/10.1145/3620678.3624663

socc23

The Gap Between Serverless Research and Real-world Systems

https://ipads.se.sjtu.edu.cn/zh/publications/LiuSoCC23.pdf

socc23

Ekko: A Large-Scale Deep Learning Recommender System with Low-Latency Model Update

osdi22

ORION: Optimized Execution Latency for Serverless DAGs

https://www.usenix.org/conference/osdi22/presentation/mahgoub

distribute system
serverless
osdi22