2019-02-12

Dundee: Petr, J-m, Jason, Simon, Riad, Dom, Mark, June,

Remote: Josh, Will, Frances, Wilma, Andreas, David, Melissa

Start: 2:00 pm

1. Accepting minutes from last meeting

2. Project Status

(2-3 minutes each)

• Data-In

  • Dom: More HPA EBI uploads

    • Someone other than Dom will import the next HPA dataset.

      Maybe.

  • David: BF 6 release in progress, official release next Monday

• Data-Out

  • Will: Iviewer: ROI pagination work. When ready we will release

    0.7.0

    • Tiled-based loading is impractical due to lack of spatial

      querying

    • ROI table pagination but it’s complicated with big images

    • See

      https://idr.openmicroscopy.org/webclient/?show=project-503 for IDR ROI example

  • Build:

    • JM: Slow progress on ice 37:

      • slice generated.

      • Upgraded path.

      • Not all Java classes are generated

      • May not make internal 5.5 milestone

    • Mark: Slow progress on indexer issue.

      • Simple indexer replacement works only by avoiding

        Spring: manual session and transaction management, no use of services, so best considered a short-term workaround. Needs more work to properly index data as done by previous indexer but lower technical risk than finally figuring out Spring issue.

    • Josh: How much would it cost to move to ElasticSearch?

      • Chris: depends on whether we replicate existing indexing

        or do something better, JSON only, how would it be deployed and integrated? Needs investigation.

• Glencoe

  • Chris: All required backbone services are on Hazelcast, async

    work ongoing. Something’s broken with 5.4.10.

• Community

  • Josh: Sysadmin fires

    • GPFS outage over weekend

    • Artifactory delete problems as usual

    • Outage caused by possible spidering of image downloads

    • Current workload may be untenable

  • HCA Report ([Breakout

    summaries</u>](https://docs.google.com/presentation/d/1xpiochfjuBQOGTvNo4K1_jpsIZ3fhr8Wiw4-rbEUl5Q/edit#slide=id.g4eb0ccbed8_0_16))

    • Jason: single cell sequencing, data is hosted on AWS and

      GCP.

      • Problem: Data egress costs are too expensive to download

        so all processing has to be done remotely

      • Could be worse.

    • Josh: Issues identified with imaging:

      • Ingest: 1000 file limit

      • Download: directories stripped

      • API: S3 access of “blobs”

      • Problems with lack of deduplication when downloading

    • General interest; dataset added over the weekend

    • Next meeting in June

  • Learning Upgrade?

    • Jason: Used by a surprisingly number of courses and people.

      Upgrade requires funding.

  • Dundee OMERO Workshops

    • Proposed dates for Users Training

      • April 8-9; June 10-11.

    • Registration?

    • Announcements & Registration (google form and website

      updates)

  • Training

    • Materials:

      http://f1000research.com

1. AOB

(5 min. max; tech. Discussion should be highlighted to relevant people and rescheduled)

• F1000 Gateways?

1. Main Topic

(20-25 minutes plus 15 minutes questions max)