MASSACHUSETTS INSTITUTE OF TECHNOLOGY HAYSTACK OBSERVATORY 4 June 2002

advertisement
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
HAYSTACK OBSERVATORY
WESTFORD, MASSACHUSETTS 01886
Telephone:
Fax:
978-692-4764
781-981-0590
4 June 2002
TO:
Distribution
FROM:
Alan R. Whitney
SUBJECT:
29 May 2002 e-VLBI telecon summary
Attendees:
Lee Foster, Pat Gary, Chuck Kodak, Kevin Kranacs, Paul Lang, Bill Wildes – GSFC
Steve Bernstein, Lorraine Prior, Peter Schultz – Lincoln Laboratory
Tom Lehman – ISI-E
Richard Crowley, Kevin Dudevoir, Hans Hinteregger, Mike Titus, Alan Whitney – Haystack
Observatory
This telecon is one of an ongoing series of telecons to prepare for gigabit/sec e-VLBI demonstrations
between NASA GSFC and MIT Haystack Observatory using a combination of network facilities
including all or part of Glownet, Bossnet, ISI-E, SuperNet, Max and GSFC/HECN.
Status Reports
The attached figures of the e-VLBI path have been updated to reflect current status and are pretty
much self-explanatory; critical status items are indicated in red. In addition, the following comments
are relevant:
− Bossnet problems
•
Peter reported that the Bossnet OC-48 link is still down (since 30 April). LL is
debugging, but problem has not been fully isolated; optical power received at D.C. from
Baltimore is 15dB down. Link was up for a few minutes 29 May but then down again.
People must be sent to debug the bad optical segment; hopefully this will happen within
next week. Good possibility that problem is in Quest fiber between Baltimore and D.C.
•
Tom reported that, before Bossnet went down on 30 April, he observed single-stream TCP
from ISI-E to LL at ~900 Mbps, but slow in other direction. When link was up briefly a
week ago, the opposite was observed; could send from LL to ISI-E at ~900, but reverse
was slow. Hopefully, once the Bossnet problem is fully resolved, we will be able to
sustain ~900 Mbps in both directions.
•
Tom can provide someone in D.C. to assist in the Bossnet debugging; tentatively
scheduled for Tuesday, 04 June 02.
minutes13.doc 6/5/2002
1
•
Tom has heard that someone might have been investigating the placing of some
equipment in the Bossnet rack in Baltimore in preparation for FC02(?) around 30 April;
details unknown, but might be worthwhile to investigate in relation to current Bossnet
problem. Jerry Sobieski at MAX should know the details.
− Tom reported that additional test workstations (G7A and G7B) have been added at ISI-E,
along with another Summit 5i (G4).
− G4 Mac system purchased by Bill Wildes has been deployed to GGAO (at L6) and named
‘pluto’; has been tested to confirm K7/L1 link from GSFC Bldg 28 to GGAO.
− Thanks to several people, a fairly complete detailed list of test workstations has now been
assembled. Alan will distribute this information via e-mail to e-VLBI group; will include
skeleton information on Mark 5 systems at A1, B1, B3 and L4.
− Lorraine reported that Lincoln radar is back on-line on Glownet, so e-VLBI must return to
scheduled usage, as in the past.
− Juniper M160 (H7) has still not been configured to use path H9. Pat and Tom will contact
Jerry Sobieski to try to get this resolved.
Performance Testing
Tom reported significantly better performance from ISI-E (G7) to Superglide (E5A) at LL than from
ISI-E (G7) to the Dell Poweredge 500C (E5), also at LL. The reason for the difference is currently
unknown. Tom is actively involved in a study to determine why data-transfer performance is so
variable among various platforms between same network nodes. Variables include motherboards,
PCI chip sets, interrupt handling, CPU’s, NIC’s, OS kernel, etc.
Kevin would like to get OS tuning parameters of all test systems in order to try to understand
performances differences.
Steve reminded everyone that the Juniper router at E7 also services traffic from MIT campus over
which we have no control and which shares with e-VLBI on OC-48 link F1.
Other
Alan reported that Haystack is preparing a proposal to NSF in collaboration with MIT Laboratory for
Computing Science and MIT LL. The proposal includes developing special protocols for e-VLBI to
better utilize network facilities, as well as infrastructure building and testing for e-VLBI.
Pat reported that GSFC is testing ‘class-of-service’ features with both Extreme and Cisco switches.
Extreme switches were easy to configure and use; Cisco switches were very difficult to configure and
could not be made functional – a trouble report has been filed with Cisco.
Action Items
Alan: Distribute workstation configuration information to group.
Steve: Keep tabs on status of Bossnet and notify group when problem is fixed.
Pat/Tom: Work with Jerry Sobieski to configure Juniper M160 at H7.
minutes13.doc 6/5/2002
2
Next telecon
Next telecon will be Mon, 24 June 2002 at 2 pm.
xc:
Steve Bernstein, LL
Jim Calvin, LL
Lorraine Prior, LL
Leslie Weiner, LL
Herbert Durbeck, GSFC
Lee Foster, GSFC
Pat Gary, GSFC
Chuck Kodak, GSFC
Kevin Kranacs, GSFC
Paul Lang, GSFC
Aruna Muppalla, GSFC
Bill Wildes, GSFC
Dan Magorian, UMCP
Tom Lehman, ISI
Jerry Sobieski, Max
Richard Crowley, Haystack
Kevin Dudevoir, Haystack
Hans Hinteregger, Haystack
Arthur Niell, Haystack
Joe Salah, Haystack
minutes13.doc 6/5/2002
3
Westford Antenna
Millstone RPE Bldg (Lorraine)
Dedicated no scheduling
GigE
WS/
Mark 5
Summit
1i
1
A2
A1
GigE
8
Alpine
A4
3804
A3
Dedicated no scheduling
28-1A
14
B1
Summit
15
100 Mb
28-1B
B8
B4
B6
B7
Hub
B10
B9
To HS switch
B3
Corr
Cntl
Cmptr
GlowNet
Congress St
Boston
(~50 km)
Amp
Cisco
1500
Connection made for e-VLBI
E3
GigE
Alpine
3804
E2
E1
D2
Lincoln Lab (Lorraine/Steve/Jim)
Schedule to be
coordinated with LL
E4A
Test
WS
From ISI-E
(’Superglide’)
GigE
E6
E4
GigE
Test
WS
Cisco
1500
3l (2 GigE + 1 OC-48)
GlowNet
(~100 km)
D1
Bossnet (Steve/Jim)
ISI-E
Scheduling on
Bossnet calendar
E7
Juniper OC-48
1310
M20
E8
GigE
E5
(Shared with LL on
scheduled basis)
GigE
B11
D3
WDM
E5A
WDM
GigE
5i
13
WS/
Mark 5
C3
16
GigE
GigE
Dell Poweredge 2500
GigE
B5
B2
LL
C2
Haystack Correlator
WS/
Mark 5
Schedule to be
coordinated with LL
C1
MIT
Campus
WDM
F1
E9
155x
Amp
Amp
F2
Bossnet currently down
Other users (no control)
To be upgraded to OC-48
Dell Poweredge 500SC
(provided by Haystack)
Figure 1: e-VLBI Path - Haystack to ISI-E
evlbi011.cdr
04 June 02
ISI-E (Tom)
Test
WS
Scheduling on
Bossnet calendar
Summit
5i
GE
G6
G7
GE
G6A
G5
Test
WS
Test
WS
G7A
G7B
Tuner
Bossnet
G1
G6B
OC-48
Summit
5i
G4
Switch
GigE
H2
GigE
G9
G3
G11
OC-48
Borrowed OC-48
interface - may not
be able to keep
G10
H4
No jumbo frames
Juniper
M160
H8
GigE
H9
Primary e-VLBI path
Not yet configured for H9 path
J3
Test
WS?
J2
H3
H6
Other users ( no control)
Cisco
4912
Switch
GigE
Summit
5i
H10
GigE
J5
J4
WDM
GigE
on loan from GSFC
H11
2 GigE
LuxN
J6
J1
Need details
OC-48
Juniper
M160
MAX/UMCP (Jerry/Dan)
GigE
No scheduling;
sufficient capacity
for e-VLBI
Backup plan
Juniper
M40
H7
MAX performance node;
need details
Test
WS?
H5
H1
G12
GigE
GigE
Trans-
ponder
G2
GE
G6C
G8
OC-48
WDM
MAX @ ISI-E (Jerry)
LX (~14km)
GGAO/Bldg 201 (Pat/Chuck)
GSFC/Bldg 28 (Pat/Paul/Kevin/Bill F.)
GigE
K2
WDM
Will try to schedule
either K2 or K3
exclusively for
e-VLBI experiment
G4 Mac with
Yellowdog Linux
K3
K1
K6
Test
WS
GGAO Antenna Trailer (Bill W./Chuck)
K4
Summit
5i
GigE
LX (~7.25km)
K7
1310
Dedicated - no
scheduling necessary
GigE Summit
L1
5i
(~500m)
L2
K5
Test
WS
GigE
L3
L5
K6A
Windows PC
(Lee Foster)
G4 Mac
SX
Mark 5
L4
Test
WS
L6
evlbi012.cdr
Figure 2: e-VLBI Path - ISI-E to GSFC/GGAO
04 June 02
Download