Top New Best Ask Show Jobs

1a1a11a

81 karma · joined March 10, 2016

submissions comments

Open Source Agent Traces with 206B tokens and 1M toolcalls

huggingface.co·7 pts·1a1a11a·

A Year in LLM Serving: Workload Evolution, Caching and Load-Balancing

arxiv.org·4 pts·1a1a11a·

Learning Augmented Heuristics

systems.seas.harvard.edu·2 pts·1a1a11a·

Burstiness is all you need for LLM serving

systems.seas.harvard.edu·1 pts·1a1a11a·

Why is the storage cluster used to train Llama 3 so slow?

scontent.fagc1-2.fna.fbcdn.net·2 pts·1a1a11a·

FIFO queues are all you need for cache eviction

s3fifo.com·6 pts·1a1a11a·

FIFO can be Better than LRU [pdf]

jasony.me·93 pts·1a1a11a·

Segcache: A memory-efficient, scalable cache for small objects with TTL

twitter.github.io·28 pts·1a1a11a·

A large scale analysis of hundreds of in-memory cache clusters at Twitter

usenix.org·2 pts·1a1a11a·