知识卡片

普通读书笔记卡 · 2313.e.5.f.11.a.2.b.3

合并流不等待全量

理解等级:中级

合并器把多个输入流统一成一个供下游处理的流,例如多个代码仓库共享构建基础设施。它减少每个来源单独部署处理系统的成本,却不意味着已收齐某一批次结果;要等全量后再统计,需要屏障而非普通合并。

这张卡片看不懂?

已有反馈 · 0 / 10 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • Chapter 12. Event-Driven Batch Processing · 查看原书依据

      A merger is the opposite of a copier; the job of a merger is to take two different work queues and turn them into a single work queue. Suppose, for example, that you have a large number of different source repositories all adding new commits at the same

    参考来源

    - 位置:《Designing Distributed Systems, 2nd Edition》第12章《Chapter 12. Event-Driven Batch Processing》(来源项目 book-sources;源文件:extracted/Designing Distributed Systems, 2nd Edition/_epub-src/OEBPS/ch12.html) - 结论依据:原文多仓库构建例子定义merger;其与join的区别在后章继续展开。 - 原始内容:A merger is the opposite of a copier; the job of a merger is to take two different work queues and turn them into a single work queue. Suppose, for example, that you have a large number of different source repositories all adding new commits at the same