VEGAS Pulsar Project Meeting - Sprint Review: 2017 July 5th 2:00 - 3:00pm ET

Present

  • Laura, Joe, Ryan, Jason, Luke, Randy, Ryan

Sprint Story Overview

  • VE-240: Move 10Gb Ethernet switch from Beamformer to test environment (IN PROGRESS)
  • VE-253: Test VPM modes with new manager version (IN PROGRESS)
  • VE-265: Resolve Astrid and config tool issue (COMPLETE)
  • VE-266: Complete manager/matrix integration and development testing (COMPLETE)
  • VE-256: Test 32-channel BOF files (IN PROGRESS)
  • VE-222: Upgrade most processing scripts to handle multi-bank data with overlapping frequency coverage (IN PROGRESS)
  • VE-267: Regression testing of release candidate 1 (IN PROGRESS)
  • VE-271: Upgrade BOF files (IN PROGRESS)
  • VE-264: Investigate bug related to spectral line matrix base line (COMPLETE)
  • VE-209: Complete installation of release candidate on reg-test environment (COMPLETE)

Story Details

VE-240: Move 10Gb Ethernet switch from Beamformer to test environment

  • Switch has been moved but not connected to machines yet.

VE-253: Test VPM modes with new manager version

  • Ray and Ryan needs to talk about the proper use of some of the coordinator/manager parameters.

VE-265: Resolve Astrid and config tool issue

  • Had to resolve some naming conventions for bank and mode names. This has now been fixed.

VE-266: Complete manager/matrix integration and development testing

  • Interface has been completed for spectral line modes. There is one config tool parameter that can be switched to run with the matrix instead of vegas_hpc. We should test this at some point but there are some other higher priority items to get to first.

VE-256: Test 32-channel BOF files

  • Ryan tried to test these using Justin's version of the manager and config tool on the real GBT system, but the manager and config tool had some hardcoded limits on the allowed number of channels, which could not be lower than 64. Justin was able to resolve some of these conflicts but Ryan was not able to get a scan to successfully run before end of maintenance time. A conflict in the DDLs, which seems to have been caused by a BTL version of the coordinator, prevented additional testing during the previous sprint. Justin may need to make some more modifications to enable scans in those mode.

VE-222: Upgrade most processing scripts to handle multi-bank data with overlapping frequency coverage

  • Ryan made some progress on this but did not finish the script.

VE-267: Regression testing of release candidate 1

  • After some additional debugging by Ryan, Joe, and Ray, Ryan was able to take data in all spectral line mode scans using the regression test environment (vegas-hpc10). Ryan took 'on/off' scans in all spectral line modes with the noise source connected and at first glance there are no anomalies. Ryan believes we can move on to full/final spectral line regression testing. The pulsar modes are not yet configuring properly. This seems to be due in part to a communication problem between the coordinator and manager, i.e. the coordinator believes a certain parameter should have a valid value, but the parameter is not changing in the manager. This requires more work to fix and has prevented more detailed tests of the pulsar modes on the regression test environment.

VE-271: Upgrade BOF files

  • Ryan tested all available BOF files with the new PFB designs, except the 32-channel modes. Most seemed to work as expected, but the c0800x2048 bandpass shape shows many artifacts and is qualitatively incorrect. This requires more work to fully characterize and debug. Note that there was a lot of packet loss in the c1500x2048 and c1500x4096 modes, which is not entirely unexpected.
  • Done and done.

VE-209: Complete installation of release candidate on reg-test environment

  • Ryan successfully used this release candidate for spectral line tests.

VEGAS Pulsar Project Meeting - Sprint Planning: 2017 June 19th 2:00 - 3:00pm ET

Note: The team's "velocity" has been consistent at 11--13 points per sprint. We should aim for 12 points per sprint for planning purposes.

Backlog

  • Complete testing of new BOF files
  • Post-processing scripts

Critical Issues

  • Complete spectral line regression tests
  • Fix remaining pulsar mode bugs in release candidate 1

New Stories

  • VE-276: Connect 10Gb Ethernet switch for test environment
  • VE-279: Pulsar mode debugging
  • VE-278: Continue Regression Test Preparation of RC1
  • VE-277: Test 32-channel coherent dedispersion BOF files
  • VE-280: Investigate c0800x2048 bandpass issues
  • VE-283: Start work on pulsar modes for matrix (R3)
  • VE-281: Continue testing BOF files
  • VE-282: Work with Justin on Manager
-- RyanLynch - 2017-07-05
Topic revision: r2 - 2017-07-17, RyanLynch
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback