MASSACHUSETTS INSTITUTE OF TECHNOLOGY
HAYSTACK OBSERVATORY
WESTFORD, MASSACHUSETTS 01886
Telephone:
Fax:
978-692-4764
781-981-0590
3 October 2002
TO:
Distribution
FROM:
Alan R. Whitney
SUBJECT:
2 October 2002 e-VLBI telecon summary
Attendees:
Bill Fink, Lee Foster, Pat Gary, Chuck Kodak, Kevin Kranacs – GSFC
Steve Bernstein, Lorraine Prior, Peter Schultz – MIT Lincoln Labs
Kevin Dudevoir, Hans Hinteregger, Arthur Niell, 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.
ACTION ITEMS ARE HIGHLIGHTED IN RED.
Status Reports
The attached figures of the e-VLBI path have been updated to reflect current status and are pretty
much self-explanatory. The following status updates were reported:
− Peter reported that Bossnet has been operational for last couple of weeks, though it is still
operating GigE. OC-48 was tried, but error rates were high and GigE was re-enabled. Tom
reported that current link from ISI-E to Eckington is from Verizon, but will go away
sometime in the next few weeks and be replaced by a higher quality fiber from Qwest that
will go directly from ISI-E to Eckington and connect to Bossnet (expected to be in place by
end of October). This new fiber should allow Bossnet to return to OC-48 and support
proposed 10GigE tests.
− On Tuesday, 9 Oct, Haystack will attempt an e-VLBI experiment with Kashima, Japan. Data
will be transmitted via Abilene and GEMnet (across Pacific Ocean) at ~20 Mbps. This is
believed to be the first attempt at intercontental e-VLBI.
− Tom reported a bad fiber was found connecting College Park (MAX) to ISI-E. This has been
fixed and will be tested with help from Bill Fink.
− Tom also reported that, as part of infrastructure upgrade, a dedicated OC-48 connection from
Bossnet to Abilene will be put in place for e-VLBI work.
minutes19.doc 10/3/2002
1
− Alan reported that two Mark 5 systems are now in place at both Westford and GGAO. A first
attempt at e-VLBI was made a few days ago, but an apparent problem in one of the Mark 5
systems at GGAO prevented success. That Mark 5 system has now been replaced and a new
attempt is planned shortly. Using two Mark 5 systems, we expect to achieve a data transfer
rate of ~800 Mbps. We will keep everyone posted on the status of these tests.
− Pat suggested that future planning needs to be done to continue to expand e-VLBI after the
current demonstrations are concluded. Alan indicated that a recent grant from NSF will allow
Haystack Observatory to continue to develop e-VLBI and that the ultimate goal is to create a
global e-VLBI network. Within NASA, Bill Wildes is very interested in promoting e-VLBI
and is a natural contact point and ally. Also, Dr. John LaBrecque at NASA HQ, manager of
the NASA Solid Earth and Natural Hazards Program, is also very interested in pursuing eVLBI. Current annual global geodetic VLBI observations are typically ~150 24-hr sessions
with 6-8 antennas each. There is desire to expand these observations and e-VLBI may help to
lower costs and increase usefulness. GGAO would be a good test platform for continuing eVLBI development efforts. We will pursue this topic further when Bill Wildes returns from
vacation.
− Alan reported that that USNO plans to lease a dark fiber from GWU to USNO. USNO will
then connect to the GigaPOP at GWU. This will allow e-VLBI data to be transmitted to
USNO for correlation on their Mark 4 correlator system.
− Pat asked about the work to create an e-VLBI demonstration connection to Hawaii. Alan
reported that Bill Wildes is taking the lead in this effort, with participation by Kokee
Observatory and Pacific Missile Range Facility (PMRF) on Kauau, and well as U. of Hawaii
on Oahu and Haystack Observatory. Tentative plan is to bring data through DREN network
to MAX, then Bossnet to Haystack. Goal is late 2002/early 2003 for demonstration. Alan
will distribute a preliminary diagram of the path prepared by Hans Hinteregger. Pat has
created a path diagram similar to the diagrams we have created for this project showing how
this path can come into the H7 router on Figure 2; Pat and Alan will distribute this material to
the project group (Pat has already done so). It is suggested that Pat be included in the
continuing preparations for this test since GGAO may be a good partner antenna for early
tests of the Kokee link.
Security
Responding to questions by Kevin D., a lengthy discussion regarding security concerns arose
involving multi-homed computers that connect to Bossnet (140.x.x.x). Lorraine made it clear that
Haystack’s use of the 140 network requires that no machine be connected to both the ‘140’ network
and any other non-secure network (e.g. Haystack’s internal network, in particular) at the same time.
In order for the e-VLBI experiments to move forward, however, it is necessary for the Mark 5 data
systems to be connected to Haystack’s internal network during data-taking and data processing;
however, they will not be connected to the 140 network during this time. They will only be
connected to the 140 network during data transfer activities, during which time they will be
disconnected from the local Haystack network. Haystack will abide by these rules.
minutes19.doc 10/3/2002
2
Alan will update network diagrams to show local-LAN-connected computers that must be connected
to the Mark 5 systems during data acquisition and correlation.
Lorraine and Steve will review the current security situation and consider alternate connections and
schemes that might be adopted in the future to ease or eliminate some or all of these constraints.
Anne Gorczyca needs to be kept in the loop in these discussions.
Report Outline
A draft outline for a proposed project report was distributed by Alan (see minutes of 16 Sep 02
telecon). After some discussion, the outline was modified as follows:
Executive Summary - Alan
Introduction - Alan
What is VLBI? - Alan
Why e-VLBI? - Alan
Goals of the 1-Gbps e-VLBI Demonstration - Alan
Elements of the 1-Gbps Demonstration
Mark 5 VLBI Data System - Alan
Glownet/Bossnet – Steve/Lorraine/Peter
ISI-E/Max – Tom/Jerry
GSFC Networks – Pat/Bill F./Paul/Kevin K./Dan
Haystack/GGAO – Kevin D./Alan/Bill W.
Testing and Performance – Tom/Bill F./Kevin K./Kevin D.
Results of e-VLBI Demonstration – Alan
Recommendations for further work/Future Directions in e-VLBI - All
Summary - Alan
Goal is to have first draft by end Nov 2002 and final report ready to submit for publication by end
Dec 2002. To be included as well: network diagrams, performance graphs, and e-VLBI results.
Everyone is asked to distribute draft material for comment; Alan will act as editor and badgerer. We
expect final report will be 10-15 pages in length.
Next telecon
Next telecon is scheduled for Wed, 23 October 2002 at 2 pm EDT.
Aruna Muppalla, GSFC
xc:
Steve Bernstein, LL
Bill Wildes, GSFC
Jim Calvin, LL
Dan Magorian, UMCP
Lorraine Prior, LL
Tom Lehman, ISI
Leslie Weiner, LL
Jerry Sobieski, Max
Herbert Durbeck, GSFC
Richard Crowley, Haystack
Bill Fink, GSFC
Kevin Dudevoir, Haystack
Lee Foster, GSFC
Hans Hinteregger, Haystack
Pat Gary, GSFC
Arthur Niell, Haystack
Chuck Kodak, GSFC
Joe Salah, Haystack
Kevin Kranacs, GSFC
Paul Lang, GSFC
minutes19.doc 10/3/2002
3
Req’d during data acq
A0
Mk5 Cntl
Cmptr
A1
Westford Antenna
Millstone RPE Bldg (Lorraine)
Mark 5
A2
SX
1
Summit
1i
GigE
SX
Westford
LAN
Mark 5
Dedicated no scheduling
A2A
Alpine
A4
3804
Haystack Correlator
Internet
B1
Haystack
LAN
Mark 5
SX
28-1A
B2
GigE
B5
B2A
14
SX 28-1A
RJ-45
13
2
B4A
1
Mark 5
B3A Mark 5
LL
C2
Dedicated no scheduling
B1A
Corr
Cntl
Cmptr
(Shared with LL on
scheduled basis)
C3
16
Summit
WDM
GigE
5i
15
B6
Cisco
1500
GigE
RJ-45
B11
B3
Test
WS
Req’d during data processing
B4
Congress St
Boston
(~50 km)
Amp
D3
WDM
E6
E3
GigE
E2
E4A
Test
WS
GlowNet
Alpine
3804
E4
GigE
Test
WS
Bossnet (Peter/Steve)
ISI-E
Scheduling on
Bossnet calendar
GigE
Juniper
M20
1310
E7
E8
GigE
E5
(~100 km)
D1
Shared with MIT traffic
Cisco
1500
E1
D2
Lincoln Lab (Lorraine/Steve/Jim)
Schedule to be
coordinated with LL
3l (2 GigE + 1 OC-48)
Dell Poweredge 2500
(’evlibhay’)
GlowNet
E5A
GigE
8
A3
A1A
Schedule to be
coordinated with LL
C1
MIT
Campus
WDM
F1
E9
Other users (no control)
To be upgraded to OC-48
155x
Amp
Amp
F2
Bossnet currently running GigE
pending investigation and repair
(’superglide’)
(’evlbill’)
From ISI-E
Dell Poweredge 500SC
(provided by Haystack)
Figure 1: e-VLBI Path - Haystack to ISI-E
evlbi011.cdr
02 Oct 02
ISI-E (Tom)
Test
WS
Scheduling on
Bossnet calendar
Test
WS
G5
G7B
G2
GE
G6B
(’kame’)
G8
OC-48
WDM
G6
Summit
5i
(’omega’)
Tuner
Bossnet
GE
G7
MAX @ ISI-E (Jerry)
G6C
G1
OC-48
Switch
GigE
G3
H5
H2
GigE
H4
G11
OC-48
Borrowed OC-48
interface - may not
be able to keep
H3
Juniper
M160
DREN
GigE
H8
GigE
H9
Primary e-VLBI path
J3
Test
WS?
J2
Other users ( no control)
Cisco
4912
Switch
GigE
Summit
5i
H10
GigE
J4
J5
WDM
GigE
on loan from GSFC
H11
2 GigE l’s
LuxN
J6
J1
Need details
Needs configuration
H6
MAX/UMCP (Jerry/Dan)
No jumbo frames
No scheduling;
sufficient capacity
for e-VLBI
OC-48
Juniper
M160
Frame relay encapsulation
G10
H7
MAX performance node;
need details
Test
WS?
Backup plan
Juniper
M40
G9
H1
G12
GigE
GigE
Trans-
ponder
Summit
5i
G4
LX (~14km)
GSFC/Bldg 28 (Pat/Paul/Kevin/Bill F.)
K4
K1
WDM
GigE
K2
GigE LX (~7.25km)
1310
K7
Summit
5i
K3
K5A
K5
G4 Mac with
Yellowdog Linux
(’clifford’)
GGAO/Bldg 201 (Pat/Chuck)
Test
WS
Test
WS
K6
K6A
K5B
Cisco
12016
K6B
GGAO Antenna Trailer (Bill W./Chuck)
GigE Summit L3
GigE
L1
5i
L5
G4 Mac
(’pluto’)
Windows PC
(’xly’)
RJ-45
Test
WS
Internet
L4
L2
Dedicated - no
scheduling necessary
(~500m)
RJ-45
Mark 5
NASA
LAN
L3A
RJ-45
L4A
Mark 5
L6
Mark 5
Cntl
Cmptr
L5
Req’d during data acq
evlbi012.cdr
Figure 2: e-VLBI Path - ISI-E to GSFC/GGAO
02 Oct 02