Uploaded image for project: 'Hadoop HDFS'
  1. Hadoop HDFS
  2. HDFS-14745

Backport HDFS persistent memory read cache support to branch-3.1

    XMLWordPrintableJSON

Details

    • New Feature
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • None
    • 3.1.4
    • None
    • Hide
      Non-volatile storage class memory (SCM, also known as persistent memory) is supported in HDFS cache. To enable SCM cache, user just needs to configure SCM volume for property “dfs.datanode.cache.pmem.dirs” in hdfs-site.xml. And all HDFS cache directives keep unchanged. There are two implementations for HDFS SCM Cache, one is pure java code implementation and the other is native PMDK based implementation. The latter implementation can bring user better performance gain in cache write and cache read. If PMDK native libs could be loaded, it will use PMDK based implementation otherwise it will fallback to java code implementation. To enable PMDK based implementation, user should install PMDK library by referring to the official site http://pmem.io/. Then, build Hadoop with PMDK support by referring to "PMDK library build options" section in `BUILDING.txt` in the source code. If multiple SCM volumes are configured, a round-robin policy is used to select an available volume for caching a block. Consistent with DRAM cache, SCM cache also has no cache eviction mechanism. When DataNode receives a data read request from a client, if the corresponding block is cached into SCM, DataNode will instantiate an InputStream with the block location path on SCM (pure java implementation) or cache address on SCM (PMDK based implementation). Once the InputStream is created, DataNode will send the cached data to the client. Please refer "Centralized Cache Management" guide for more details.
      Show
      Non-volatile storage class memory (SCM, also known as persistent memory) is supported in HDFS cache. To enable SCM cache, user just needs to configure SCM volume for property “dfs.datanode.cache.pmem.dirs” in hdfs-site.xml. And all HDFS cache directives keep unchanged. There are two implementations for HDFS SCM Cache, one is pure java code implementation and the other is native PMDK based implementation. The latter implementation can bring user better performance gain in cache write and cache read. If PMDK native libs could be loaded, it will use PMDK based implementation otherwise it will fallback to java code implementation. To enable PMDK based implementation, user should install PMDK library by referring to the official site http://pmem.io/ . Then, build Hadoop with PMDK support by referring to "PMDK library build options" section in `BUILDING.txt` in the source code. If multiple SCM volumes are configured, a round-robin policy is used to select an available volume for caching a block. Consistent with DRAM cache, SCM cache also has no cache eviction mechanism. When DataNode receives a data read request from a client, if the corresponding block is cached into SCM, DataNode will instantiate an InputStream with the block location path on SCM (pure java implementation) or cache address on SCM (PMDK based implementation). Once the InputStream is created, DataNode will send the cached data to the client. Please refer "Centralized Cache Management" guide for more details.

    Description

      We are proposing to backport the patches for HDFS-13762, HDFS persistent memory read cache support, to branch-3.1.

      Attachments

        1. HDFS-14745-branch-3.1-003.patch
          149 kB
          Feilong He
        2. HDFS-14745-branch-3.1-002.patch
          147 kB
          Feilong He
        3. HDFS-14745-branch-3.1-001.patch
          142 kB
          Feilong He
        4. HDFS-14745-branch-3.1-000.patch
          142 kB
          Feilong He

        Issue Links

          is related to

          New Feature - A new feature of the product, which has yet to be developed. HDFS-13762 Support non-volatile storage class memory(SCM) in HDFS cache directives

          • Major - Major loss of function.
          • Resolved
          relates to

          New Feature - A new feature of the product, which has yet to be developed. HDFS-14905 Backport HDFS persistent memory read cache support to branch-3.2

          • Major - Major loss of function.
          • Resolved

          Activity

            People

              PhiloHe Feilong He
              PhiloHe Feilong He
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: