Uploaded image for project: 'Fabric'
  1. Fabric
  2. FAB-16281

Performance degradation with large ledger

    Details

    • Type: Story
    • Status: To Do (View Workflow)
    • Priority: Medium
    • Resolution: Unresolved
    • Affects Version/s: None
    • Fix Version/s: None
    • Component/s: None
    • Labels:
      None
    • SDK Impact:
      No
    • System Test Impact:
      Yes
    • Documentation Impact:
      Yes

      Description

      When doing the duplicate transaction checking for each transaction, it will query whether the transaction id exists in the indexDB.

      With large ledger, there will be lots of existing transactions information in the indexDB, and the performance will go down. Since this is per-transaction behavior, it will affect the performance nonNegligible.

      This issue will become worse with accumulating the transactions over time.

       

      There may be several workarounds:

      1) Enhance indexDB (which uses levelDB now) by using better DB in terms of large-scale.

      2) Design specific data structure to enhance this duplication checking.

      3) Re-design the existing transaction id generation mechanism, e.g., add timestamp to allow storing with separate smaller DB size.

       

        Attachments

          Activity

            People

            • Assignee:
              Unassigned
              Reporter:
              baohua Baohua Yang
            • Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated:

                Git Source Code