Uploaded image for project: 'Samza'
  1. Samza
  2. SAMZA-863

Support multi-threading in samza tasks

    XMLWordPrintableJSON

Details

    • New Feature
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 0.11.0
    • 0.11.0
    • None
    • None

    Description

      Currently a samza container executes the tasks sequentially in a single thread. For example, we have message 1 and 2 in the pending queue for task 1 and task 2. Task 1 will process message 1, and until its completion task 2 can process message 2. If we want to handle more messages in parallel, we have to increase the container count, e.g. from 1 to 2 in the example.

      While this solution has been working for many CPU-bound job scenarios, we do see its drawback for IO-bound jobs.In this kind of jobs, the task makes IO/Network requests, i.e, db calls, rest calls or external service RPC calls. These IO calls significantly slow down the task processing. We can increase container number in order to parallelize the IO calls, but it results in low CPU utilization. If we can improve CPU utilization by allocating multiple contains in the same CPU core, it will still cause dramatic memory growth due to the memory being allocated for each container.

      To better scale the performance of IO-bound jobs, we are proposing to support multi-threaded processing in samza. The design proposal will come soon.

      rbs:
      https://reviews.apache.org/r/48243/: SAMZA-961: Async tasks and multithreading model

      https://reviews.apache.org/r/48213/: SAMZA-960: Make system producer thread safe

      https://reviews.apache.org/r/48182/: SAMZA-958: Make store/cache thread safe

      Attachments

        1. DESIGN-SAMZA-863-0.pdf
          367 kB
          Xinyu Liu
        2. DESIGN-SAMZA-863-1.pdf
          405 kB
          Xinyu Liu
        3. DESIGN-SAMZA-863-2.pdf
          417 kB
          Xinyu Liu
        4. DESIGN-SAMZA-863-3.pdf
          417 kB
          Xinyu Liu
        5. perf-test-results.pdf
          144 kB
          Xinyu Liu
        6. SAMZA-863.0.patch
          189 kB
          Xinyu Liu
        7. SAMZA-863.1.patch
          189 kB
          Xinyu Liu
        There are no Sub-Tasks for this issue.

        Activity

          People

            xinyu Xinyu Liu
            xinyu Xinyu Liu
            Votes:
            1 Vote for this issue
            Watchers:
            9 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: