Skip to content

怎样设定合适的cpu和内存来运行算法 #163

Closed

Description

@songlf1002

Problem Type (问题类型)

configs (配置项 / 文档相关)

Before submit

Environment (环境信息)

  • Server Version: v0.11.x
  • Computer Version: v0.x
  • Scheduler: K8s V1.xx
  • Data Size: 10 million vertices, 1 billion edges

Your Question (问题描述)

怎样设置最佳的k8s的cpu和内存大小,来运行算法,我设置的是下面这种,出现oom
workerCpu: 1
workerMemory: 1024Mi
workerInstances: 3

Vertex/Edge example (问题点 / 边数据举例)

No response

Schema [VertexLabel, EdgeLabel, IndexLabel] (元数据结构)

No response

Activity

  1. imbajin commented on Dec 13, 2021

    @imbajin
    Member

    最佳配置好像还得等等看, 你可以先试试1亿左右点边的数据集看看. 以免读取时间也过长

    因为现在是直接从 graph server 读取数据, 所以你可以先减少点数据量来试试, 后续有 hdfs 直接读取再换大数据集看看.

    另外也可以补充一下3个 worker 的资源/磁盘/网络等使用监控图 (如果有) 以及 OOM 的相关信息 (内存变化情况)

    另外, cpu 和内存不用给的这么小~ 第一次可以稍大一点试试, 如果发现使用不多再减少

  2. github-actions commented on Dec 28, 2021

    @github-actions

    Due to the lack of activity, the current issue is marked as stale and will be closed after 20 days, any update will remove the stale label

  3. coderzc commented on Jan 6, 2022

    @coderzc
    Member

    jvm 默认会将最大堆设置为容器限制的 1/4,我们将改为 1/2 https://github.com/hugegraph/hugegraph-computer/blob/master/computer-dist/Dockerfile#L4

  4. github-actions commented on Jan 22, 2022

    @github-actions

    Due to the lack of activity, the current issue is marked as stale and will be closed after 20 days, any update will remove the stale label

  5. github-actions commented on Feb 12, 2022

    @github-actions

    Due to the lack of activity, the current issue is marked as stale and will be closed after 20 days, any update will remove the stale label

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions