West Coast Spectrometer Team

22
West Coast Spectrometer Team Mark Wagner, Berkeley project manager, FPGA designer Terry Filiba, data transport: FPGA --> CPU --> GPU Suraj Gowda, boosting FFT/FPGA clock speed Glenn Jones, digital downconverter design (caltech) Guifre Molera, 10Gbit ethernet protocol, GUPPI mods Gregory Desvignes, Guppi Code modifications Simon Scott, systems integration, arriving march 26 Hong Chen, fft optimizations (bit growth, unscramber) Billy Mallard, DSP library optimizations (DSP48, etc) Andrew Siemion, galactic center pulsar application Dan Werthimer, taking credit for above work 1

description

West Coast Spectrometer Team. - PowerPoint PPT Presentation

Transcript of West Coast Spectrometer Team

Page 1: West Coast Spectrometer Team

West Coast Spectrometer Team

Mark Wagner, Berkeley project manager, FPGA designer Terry Filiba, data transport:  FPGA -->  CPU -->  GPU Suraj Gowda,  boosting FFT/FPGA clock speed Glenn Jones, digital downconverter design (caltech) Guifre Molera,  10Gbit ethernet protocol, GUPPI mods Gregory Desvignes, Guppi Code modificationsSimon Scott, systems integration, arriving march 26 Hong Chen, fft optimizations (bit growth, unscramber) Billy Mallard, DSP library optimizations  (DSP48, etc) Andrew Siemion, galactic center pulsar application      Dan Werthimer, taking credit for above work

1

Page 2: West Coast Spectrometer Team

Observing Modes: 250 MHz GPU

2

Page 3: West Coast Spectrometer Team

3

Page 4: West Coast Spectrometer Team

Overall Block Diagram

4

Page 5: West Coast Spectrometer Team
Page 6: West Coast Spectrometer Team

Roach Motel (Roach Nest) (KAT)

Page 7: West Coast Spectrometer Team

7

Page 8: West Coast Spectrometer Team

Roach I vs Roach II

• Roach I works well. Deployed at many observatories• Roach II doesn’t exist. Prototypes spring. Prod Winter?

• Roach I resources are tight, harder to get to work at high speed hard to add features, 500 MHz 8K channels won’t fit

• Roach II can use SFP+ connectors, more reliable 10Gbe connector

• Plan: develop and test using Roach I. Decide later.

8

Page 9: West Coast Spectrometer Team

3 GS/s ADC Board

9

Page 10: West Coast Spectrometer Team

10

Page 11: West Coast Spectrometer Team

Polyphase Filter BankPolyphase Filter Bank

Page 12: West Coast Spectrometer Team

FGPA Spectrometer – Mark Wagner

12

Page 13: West Coast Spectrometer Team

FPGA DDC/Packetizer (Mark Wagner) (extract sub-band(s) and send to GPU)

13

Page 14: West Coast Spectrometer Team

64 channel spectrometer

2 GHz bandwidith with xilinx place/route

Page 15: West Coast Spectrometer Team

1024 channel spectrometer, 3 GHz BW Suraj Gowda scripts for autoplacement

Page 16: West Coast Spectrometer Team

“Automated Placement for Parallelized FPGA FFTs” Suraj Gowda et al, 2011

No Placement Constraint Placement Constrained using our algorithm

Processable Bandwidth <2.4 GHz > 3 GHz

Compile time 80:19 minutes 38:22 minutes

Page 17: West Coast Spectrometer Team

x0 xx1 xx2 xx3 xx4 xx5 xx6 xx7 x

E0(z)

E1(z)

E2(z)

E3(z)

E4(z)

E5(z)

E6(z)

E7(z)

+… y3 y2 y1 y0

8*Fclk real samples per second (BW=4*Fclk)

Fclk cplx samples per second (BW=Fclk)

Existing CASPER DDC/Decimation filterQuarter band filter for 8 real inputs

Multiply by complex sinusoid

Polyphase filter components

Page 18: West Coast Spectrometer Team

Half Band DDC/Filter - Glenn Jones

x0x4 x

x1x5 x

x2x6 x

x3x7 x

E0(z)

E1(z)

E2(z)

E3(z)

… y2 y0

2*Fclk cplx samples per second (BW=2*Fclk)

+

8*Fclk real samples per second (BW=4*Fclk)

v0v4

v1v5

v2v6

v3v7

+

… y3 y1

Page 19: West Coast Spectrometer Team

SPEAD packet - FGPA 10Gbe (Guifre Molera)

Page 20: West Coast Spectrometer Team

Preliminary Design Work• Concentrating on the hard parts

– 3 GS/s sampling and PFB/FFT calculations– Heterogeneous Computing Approach

• Divide processing into front/back ends• Use FPGAs to fully process bandwidths greater than 250

MHz• Use FPGA front-ends to pre-process, split and packetize

data, then GPUs to provide fine channelization on narrower chunks

– Software Design• Adapting code from the Green Bank Ultimate Pulsar

Processing Instrument (GUPPI)

20

Page 21: West Coast Spectrometer Team

Pulsars at the Galactic Center ??100’s of pulsars predicted in the central pc none undiscovered - Macquart, Frail, Ransom,

Bower..

Map gravitational field (timing), ISM at GC, black hole spin?

Extreme scattering smears out the pulse

High Frequency Observation to minimize scattering

High Bandwidth Needed at High Frequency (low flux) 800 MHz 8 GHz

21

Page 22: West Coast Spectrometer Team

Worries• Speed of FPGA - difficult, time consuming

layouts, perhaps impossible for full Roach I chip

• Lots of modes - time consuming (design/test/software/document)

• Will everything fit in Roach I? Use Roach II?

• Might loose Mark Wagner in September22