pipelining

博客文章介绍如何通过 CyclicBarrier 将锁步（lockstep）tick 游戏循环改造为流水线（pipeline）tick 系统，只需约三行代码改动即可让读写阶段并行执行，从而将 tick 时间从读写时间之和降到最大单阶段时间，同时保留背压与超时跟踪能力。

This paper introduces techniques to manage memory peaks in training large Mixture-of-Experts models with long context lengths, including Pipelined LLEP, Ring-DTP, SCO, and OffloadStreamAdamW, which enable fixed GPU working sets and improve throughput up to 10.4x.

StreamMA introduces a streaming communication paradigm for multi-agent reasoning that pipelines intermediate results to reduce latency and improve effectiveness by leveraging more reliable early steps, outperforming baselines across benchmarks and revealing a step-level scaling law.