Time/Place

• Time: 11:00AM Eastern Time US (UTC-5)
• Dial-in Number: (712) 775-7035
  • Participant Code: 479307#
  • International numbers: Conference Call Information
  • Web Access: https://www.freeconferencecallhd.com/wp-content/themes/responsive/flashphone/flash-phone.php  

Attendees 

• Andrew Woods
• Yinlin Chen
• Colin Gross
• Danny Bernstein
• Trey Pendragon
• Longshou Situ

Agenda

1. Review last meeting actions:
   
   • Colin Gross to transfer fcrepo_perf_analysis to fcrepo4-labs
   • Andrew Woods to post test outputs to S3
   • Colin Gross to push outputs from R-scripts to GitHub
     https://github.com/fcrepo4-labs/fcrepo_perf_analysis/tree/feature/dist-upload/dist
     
   • Nick Ruest and Yinlin Chen to take above R-script output as starting point for performance test summary report
2. Review current GitHub issues: https://github.com/fcrepo4-labs/fcrepo_perf_analysis/issues
3. Infrastructure options - Research Grant 
   1. AWS Cloud Credits for Research https://aws.amazon.com/research-credits/ Rewards varies from 1 to 2 years.
4. Reproduce new performance report?
   1. If anyone what to know current Fedora 4.7 performance, we can refer him to see the performance report page.
   2. Performance and Scalability Test Plans
   3. Baseline performance on these six scenarios in a: 
      1. Select an instance type in AWS
      2. other options?
   4. An Ansible tool to install Fedora 4 on the local machine or AWS. 
      1. https://github.com/VTUL/fcrepo4-ansible
5. Define tests of known pain points, e.g.
   1. Performance of "many members"
   2. Ugliness of pairtree resources
   3. migration of PIDs from fcrepo3 to fcrepo4
   4. Migration from fcrepo3 to fcrepo4 in general and in a reasonable amount of time for 1M+ objects
   5. Loss of system dates, the migration of creation/modification times from fcrepo3 to fcrepo4
   6. Intra-repository referential integrity can be a pain point, but it has benefits too

Minutes

Previous actions

1. Updated scripts to use perf.log instead of the csv
   • GitHub feature branch contains the generated reports
   • Hydra: would like a concise report
     • version of fedora
     • configuration of fedora
     • how tests relate to real-world questions
     • what these tests mean to the fedora development team

GitHub Issues

1. Colin has it covered

Infrastructure options

1. Need estimates of storage and instances
   • sometimes need 20-TB of data, for example
   • i/o latency
     • writes to /dev/null
     • reads are based on hash of path of resource
       + Self and Yinlin to put in proposal https://aws.amazon.com/research-credits/

New performance reports

1. Running against 4.7.1
2. Waiting on open actions from fcrepo_perf
3. Need documented process for end-to-end testing and reporting
   + create new wiki page : danny

Actions

• Colin Gross will finish up current GitHub issues: https://github.com/fcrepo4-labs/fcrepo_perf_analysis/issues
• Andrew Woods and Yinlin Chen will work on AWS grant proposal, deadline 03/31/17, will have a draft for review on next meeting: https://aws.amazon.com/research-credits/ 
• Danny Bernstein and Longshou Situ will have a wiki page about a documented process for end-to-end testing and reporting