• Type: Icon: Epic Epic
    • Resolution: Done
    • Priority: Icon: Unknown Unknown
    • None
    • Affects Version/s: None
    • Component/s: BSON
    • None
    • Done
    • Raw BSON Types
    • 5
    • 10
    • 10
    • 100
    • Hide

      Engineer: Patrick Freed
      Summary: This project covers the introduction of a raw BSON document API to the Rust BSON library and the associated implementation updates to the Rust driver required to support retrieving results from the server using that API.

      2022-04-01: Project complete.


      2022-03-04: Project complete, final blog post is in review.


      2022-02-17: Increased the final cost to 10 and updated the end date to reflect the additional 2 weeks to add support for zero-copy deserialization in cursors.


      2022-02-04: Updated end date to 02/11/22 due to time off and final blog post.

      -Patrick submitted final PR to TechEmpower and it is in review.  

      -Blog post is in progress and should be published end of next week.

       


      2022-01-21: Updated final cost due to additional work for Benchmarking
      -Final benchmarking work in progress and blog post is pending.


      2022-01-04: Pending reviews from external teams, updated end date to 01/19/22.
      -Benchmarking and minor blog post remaining.


      2021-12-14: Updated target end date to 2022-12-17
      -Updated end date due to holiday time off.


      2021-11-29: Updated target end date to 2021-12-17
      -Updated end date due to additional time off and skunk works.


      2021-11-12: Updated target end date to 2021-12-03

      • Patrick was focused on higher priority tasks, not incrementing spend.
      • End date updated to reflect delay and holiday time off.

      ---- - Original user PR has been refined and merged.

      • Left to do: finish the deserialization work, appending the driver work, and writing a short blog post.
      • Feels like there's more than two weeks of engineering left, incrementing the final cost by 1 week.

      ---- - Merging the users PR has been somewhat more complex than expected, but we still think it will be possible to complete within 2 weeks (estimate still valid despite higher complexity, buffer in initial estimates are useful!)

      • Still on track for cost expenditure, baking in a 2 week buffer since it is rare to get a week of 100% focus.
      Show
      Engineer: Patrick Freed Summary: This project covers the introduction of a raw BSON document API to the Rust BSON library and the associated implementation updates to the Rust driver required to support retrieving results from the server using that API. 2022-04-01: Project complete. 2022-03-04: Project complete, final blog post is in review. 2022-02-17: Increased the final cost to 10 and updated the end date to reflect the additional 2 weeks to add support for zero-copy deserialization in cursors. 2022-02-04: Updated end date to 02/11/22 due to time off and final blog post. -Patrick submitted final PR to TechEmpower and it is in review.   -Blog post is in progress and should be published end of next week.   2022-01-21: Updated final cost due to additional work for Benchmarking -Final benchmarking work in progress and blog post is pending. 2022-01-04: Pending reviews from external teams, updated end date to 01/19/22. -Benchmarking and minor blog post remaining. 2021-12-14: Updated target end date to 2022-12-17 -Updated end date due to holiday time off. 2021-11-29: Updated target end date to 2021-12-17 -Updated end date due to additional time off and skunk works. 2021-11-12: Updated target end date to 2021-12-03 Patrick was focused on higher priority tasks, not incrementing spend. End date updated to reflect delay and holiday time off. ---- - Original user PR has been refined and merged. Left to do: finish the deserialization work, appending the driver work, and writing a short blog post. Feels like there's more than two weeks of engineering left, incrementing the final cost by 1 week. ---- - Merging the users PR has been somewhat more complex than expected, but we still think it will be possible to complete within 2 weeks (estimate still valid despite higher complexity, buffer in initial estimates are useful!) Still on track for cost expenditure, baking in a 2 week buffer since it is rare to get a week of 100% focus.

      Summary

      This epic captures the Rust driver performance improvement work of implementing Raw BSON types.

            Assignee:
            patrick.freed@mongodb.com Patrick Freed
            Reporter:
            alexander.golin@mongodb.com Alexander Golin (Inactive)
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

              Created:
              Updated:
              Resolved:
              20 weeks, 6 days