VEGAS Pulsar Project Meeting: 2015 December 7th 2:00 - 3:00pm ET

Room / Connection Details

  • GB-137 / Soc 280
  • 192.33.117.12##7144
  • 434-817-6443

Housekeeping Details

  • Main wiki page
  • gbsapp Mailing List
  • Location and regular date and time of meeting - agreed weekly on Mondays; GB-137 / Soc-280; 2-3pm ET
  • ETK codes:
    • Electronics: 432510.GB3015
    • Software: 893252.GB3015
    • Science: 432540.GB3015

Resource Allocations

BLOSS, MARTIN ALAN 0.05
BRANDT, JOSEPH JOHN 0.10
CREAGER, RAY 0.65
DEMOREST, PAUL 0.05
RICHMOND-DECKER, JUSTIN 0.20
FORD, JOHN M 0.05
MELLO, MELINDA JANE 0.05
MCCULLOUGH, RANDY L 0.25
RAY, JASON 0.55
PRESTAGE, RICHARD 0.05
LYNCH, Ryan 0.25
VAN TONDER, VEREESE 0.10

Present

  • Mark, Ryan, Richard, Randy, Jason, John, Joe, Randy, Justin

BOF file status updates

  • c1500x4096 mode
    • Joe does not think the GPU was actually being used. Need to look more closely.
      • The pre and post transpose was not being carried out on the GPU
      • Joe/Randy will recompile and Ryan will test on simulator
    • nVidia profiling
      • No update
    • guppi_daq_server version comparisons
      • No update
  • c1500x0512 mode
    • Build status
      • Randy took original c1500x0512 design under Xilinx 14.7, simply changed the file name, and was able to build with no timing errors. Previous to this it was building with timing errors under new libraries. Still working to understand this. Randy does not think there should be anything cached or carried over that would explain this.
      • Other than the file name Randy is happy to release this, but we should try to understand this behavior.
      • Ryan will begin testing this ASAP
    • Astronomical verification
      • Will try to test this with GBT hardware this week if simulator tests go well
  • c1500x2048 mode
  • Fix for nchan calculation in coherent modes
    • Ready to implemented in other modes
  • Bandpass issues in i0800x0064, 0128, 8192
    • Spent some time testing 64 channel mode
    • Tone shows up in the correct bins, but generates spurs if injected at Fs/8 or nearby
      • Spurs move with the test tone and decrease with amplitude if you move away from Fs/8
      • May be problem with MMCM or OGP calibration
        • Will try to check with real VEGAS hardware on Thursday afternoon
        • Justin will talk with Paul M about MMCM and OGP calibration
        • All snapblocks with correct names are present
  • Duplicated signal in i0800x8192
  • Low bandwidth modes
    • New filter coefficients
    • Spectrum output instead of time series
    • Implement PFB code
    • Could single GPU keep up with coherent dedispersion?

Software

  • Configtool changes
    • First tests carried out with GBT
      • Had to add new parameter for polarization mode ('total_intensity' or 'full_stokes')
        • Seems to only control the onlyI parameter. Is there any other behavior this controls?
        • This may be redundant with vegas.vpol
          • vegas.vpol calculates xx,yy (self), xx,yy,xy,yx (cross), xx (self1), yy (self2)
      • Config tool ran into error with 'tfold' parameter
        • Justin thinks he tracked that down to a typo
      • Is it OK for parfile to be an empty string for cal and search modes?
    • Will need to rebase on top of new config tool version to bring in Melinda's changes to avoid center spike
  • GBT Status shared memory parameters in PSRFITS headers
    • Set observer and source and check PSRFITS headers
    • Verify with GBT hardware
  • Switching signal generation
    • swper is not being set correctly in VEGAS
      • Request 0.04 s w/ blanking lasting 0.002 s, but get 0.0409754 s w/ blanking lasting 0.0024986 s
        • This may be related to the limits on the 'granularity' in spectral line mode
        • Don't need to know equivalents of hwexposr and spectick in coherent modes
          • Just need click tick and number of channels to program the bof to switch at the right number of samples
          • In incoherent mode, since we do accumulate, so this should work the way modes 1--3 do in VEGAS
          • Need a table with spectick parameters for different pulsar mode tints
      • Alternative is to use DCR as switching signal master
        • We know this works
        • Already done with GUPPI
        • Don't want to do this because some users have wanted to do fast switching, which VEGAS can do
    • Randy discovered that there are two values of nchan in the designs, one for switching signal generation and one for the rest of the logic
      • Need to test on srbs-hpc1
  • Matrix HPC
    • Just asks what difference between search packet and baseband packet
    • When writing baseband packet, it does a transpose and copy and when writing search packet it just does a copy
    • Ask Paul D about how these get used
    • If there are no major differences, should be able to combine incoherent and coherent net thread
      • But coherent modes overlap blocks to avoid edge artifacts in the FFT

GBT testing

  • Tested config tool changes. Will not have much remaining time for testing this week.

Wideband AP

  • Status of new filters
    • No updates

VEGAS/DIBAS-style backend for FAST

  • Documents sent to Karen
  • Karen recommends sending to Tony B

Multicasting for spectral line and pulsar modes

  • Modifications to ROACH configuration and net task must bind to a multicast group instead of a specific host
    • All the 10-gig addresses on the ROACH would be multicast groups, and config will be told which to bind to
    • Some changes to net task code (in VEGAS HPC and guppi_daq code) to initialize sockets

Date / Time of Next Meeting

  • December 14

-- RyanLynch - 2015-11-30
Topic revision: r2 - 2015-12-07, 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