llm-inference

Recent Posts
LLM 推理优化学习主干道
01 · Core Concepts
02 · 日志抽象与 Push vs Pull
03 · 存储设计：Segment、Page Cache、零拷贝、Retention、Log Compaction
05 · Replication / ISR / Leader Election
Categories
Tags
topic consumer-group kraft arithmetic-intensity compression profiling continuous-batching partition 进程管理 llm occupancy engineering HTTP rebalance offset reference group-coordinator raft bank-conflict inference consumer linking autoregressive header acks move-semantics gemm llm-inference metadata transformer LSO Linux thread DNS throughput ISR smart-pointer 侵入式链表 tiling pull
Archives
Website Info
Article Count :
24
Total Word Count :
44k
Unique Visitors :
Page Views :
Last Update :