← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 5377.e

批处理与资源编排分层

理解等级:中级

MapReduce类系统提供大规模批数据处理的执行方式;容器和虚机管理器负责在物理硬件上管理大量运行单元。前者围绕数据计算任务,后者围绕资源与部署生命周期。可组合使用但不彼此替代;本章仅列举Spark、Kubernetes和VMware,不据此推导唯一或普遍最佳实现。

相关:[[持久工作流跨越机器寿命]]。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • 宏观架构导论 · 查看原书依据

      MapReduce systems: Let us process large amounts of data as batch tasks. Apache Spark is the production-grade implementation used exclusively. Containers/VM managers: Control large collections of containers or virtual machines (VMs) deployed on top of real hardware. Most modern systems are deployed on top of container orchestration systems because they provide better control. Kubernetes is the widely used solution for containers, and VMware is widely used for VMs.

    参考来源

    - 位置:《Software Architecture and Decision-Making》第5章《宏观架构导论》(来源项目:book-sources;源文件:extracted/Software Architecture and Decision-Making Leveraging Leadership, Technology, and Product Management to Build Great Products… (Srinath Perera)/_epub-src/ch05.xhtml) - 结论依据:原文的定义、比较或例子支持“MapReduce类系统提供大规模批数据处理的执行方式;容器和虚机管理器负责在物理硬件上管理大量运行单元”;正文据此说明用途和边界。 - 原始内容:MapReduce systems: Let us process large amounts of data as batch tasks. Apache Spark is the production-grade implementation used exclusively. Containers/VM managers: Control large collections of containers or virtual machines (VMs) deployed on top of real hardware. Most modern systems are deployed on top of container orchestration systems because they provide better control. Kubernetes is the widely used solution for containers, and VMware is widely used for VMs.