Transcript Some Sample Commercial Network Costs

Goals of a New VLBI Data System
•
•
•
•
•
•
•
•
•
•
Low cost
Based primarily on unmodified COTS components
Modular, easily upgradeable
Robust operation, low maintenance cost
Easy transportability
Conformance to VLBI Standard Interface specification (VSI)
Compatibility with existing VLBI systems
Flexibility to support e-VLBI
Minimum of 1 Gbps data rate
24-hour unattended operation at 1 Gbps
Tape vs. Disc Price Comparison
7
6
5
Disc Drive
Street Prices
log($/GB)
4
1998 IBM
Disc Projection
3
c
Computer Tape
Media
2
1998 NSIC
Disc Projection
1
LTO Media
Projection
Mark IV/VLBA/K4 Media
0
S2, S3 Media
Disc industry
Projections
-1
-2
1980
1985
1990
1995
Year
2000
2005
2010
Advantages of Magnetic Discs
•
•
•
•
•
•
Readily available consumer product
Standard electrical interface
Technology improvements independent of electrical interface
Self contained; don’t have to buy expensive tape drives
Rapid random access to any data
Essentially instant synchronization on playback to correlator (no
media-wasting early starts needed)
• No headstacks to wear out or replace – ever!
Virtually maintenance free.
Mark 5 VLBI Demonstration System – March 2001
3 months start to finish!
Disk-based VLBI Data Systems
• Europe – PC/EVN
–
–
–
–
•
Developed primarily by Metsahovi group
Supports partial implementation of VSI-H specification
Uses 5 PC’s to support recording and playback at 1 Gbps
Has been used for several European experiments
Japan – K5
– Developed by CRL
– Uses 4 PC’s to support recording and playback at 512 Mbps
– Several experimental units deployed in Japan
•
Australia – Swinburne group
– Based on Apple Xserve RAID array interface
– Designed for pulsar work; plan to use for VLBI at 128 Mbps
•
U.S. – Mark 5A
– Developed at Haystack Observatory with support of international consortium
(BKG,
EVN, KVN,MPI, NASA, NRAO & USNO)
– 1 Gbps in single chassis
– Chosen by IVS and EVN to replace Mark 4 and VLBA tape systems
– ~30 systems now deployed; expect 60-70 by end 2003
Mark 5A VLBI Data System
Mark 5A Characteristics
•
•
•
•
•
•
•
•
1 Gbps recording and playback
Direct ‘plug-compatible’ replacement for VLBA or Mark 4 tape drive
‘8-pack’ disk modules for ease of disk handling
Packaged in single 5U chassis
Supports e-VLBI
Supported by NASA Field System
Supported by Mark 4 correlators (JIVE, MPI, USNO, Haystack)
Available commercially from Conduant Corp for ~$16K
(compared to Mark 4/VLBA tape drive for ~$150K)
• Plan intercompatible support for ATA serial disks
• Plan inexpensive ‘expansion chassis’
• Mark 5B (VSI-compatible) system under development;
will allow VLBA to support 1 Gbps; Mark 5A can be upgraded with
interface card replacement
Mark 5A Experience
• Daily UT1 ‘Intensive’ observations Wettzell-Hawaii have been
exclusively Mark 5 for ~10 months – almost no problems
• 15-day geodesy experiment in Oct 02 successfully recorded and
processed from Westford antenna; processed one disk set with two
missing disks!
• Several astronomy experiments, including recent mm-VLBI
experiment have successfully used Mark 5A
• Have been some teething problems, particularly learning to properly
manage disks, but believed under control
Disk-Media Status
• Hard disk price vs capacity/performance will continue to drop rapidly
- Now ~$1.00/GB, expected to drop to ~$0.50/GB by ~2005
(Mark 4/VLBA tape is ~$2.00/GB)
• 200 GB disks now available – 27 hours @ 256 Mbps unattended
• (comparable to ~5 VLBA tapes)
• 320 GB disks expected soon – 22 hours @ 512 Mbps unattended
(comparable to ~9 VLBA tapes)
• 700 GB disks expected ~2005 – 24 hours @ 1 Gbps unattended
(comparable to ~19 VLBA tapes)
Strawman Plan for VLBA
36 Mark 5 systems - ~$600K
Storage requirements:
24 hours @ 1 Gbps = 11TB
300 station-days @ 1 Gbps = 3300TB
Media costs
Rate
(Mbps)
Total
(TB)
Disk
($/GB)
Other
($/GB)
2003
256
825
1
0.15
1.0M$
2004
512
825
0.75
0.15
+0.8M$
2005
1024
1750
0.5
0.10
+1.0M$
Total
3300
~Cost
2.8M$
Total cost over 3 years ~3.4M$
VLBA correlator:
• Should be able to keep up with 512 Mbps for up to 20 stations
• Can process 1024 Mbps for up to 10 stations
e-VLBI: The Next Step
• Bandwidth growth potential for higher sensitivity
– VLBI sensitivity (SNR) proportional to square root of Bandwidth
resulting in a large increase in number of observable objects
(only alternative is bigger antennas – hugely expensive)
– e-VLBI bandwidth potential growth far exceeds recording capability
(10 Gbps per station is possible today; 100 Gbps in near future!)
• Rapid processing turnaround
– Astronomy
• Ability to study transient phenomena with feedback to steer observations
– Geodesy
• Higher-precision measurements for geophysical investigations
• Better Earth-orientation predictions, particularly UT1, important for
military and civilian navigation
Practical Advantages of ‘e-VLBI’
• Increased Reliability
– remove recording equipment out of field
– remote performance monitor & control capability in near real-time
• Lower Cost
– Automated Operation Possible
• eliminates manual handling and shipping of storage media
– Real-time or near-real-time Processing
• forestalls growth of storage-capacity requirements with bandwidth growth
– Reduce and perhaps eliminate expensive recording-media pool
(millions of $’s!)
• Avoid unexpected media-shipping interruptions and losses
Japan has led the way in e-VLBI!
• 1998-2001: Keystone project
– 4 antennas around Tokyo area connected in real-time at 256 Mbps
• 2000: Gbps ftp e-VLBI demonstration
• 2001: ATM-based 1 Gbps real-time VLBI
• 2002: 2 Gbps VLBI demonstration
Connecting the Global VLBI Array in the New Era of High-Speed Networks
8-9 April 2002
MIT Haystack Observatory, Westford, MA USA
Sponsored by:
International VLBI Service (IVS)
Global VLBI Working Group (GVWG)
MIT Haystack Observatory
With the world increasingly wired for high-speed data communications, the prospects for routine global electronic
transmission of VLBI data (dubbed ‘e-VLBI’) become brighter every day. Not only will e-VLBI help eliminate costly and
complex recording equipment, but it should eventually lead to data rates and volumes unattainable by traditional recording
equipment. This will lead to improved sensitivity, allowing new science to be explored at lower costs.
This international two-day meeting at MIT Haystack Observatory is being organized to explore the current state of highspeed astronomy data transmission, concentrating on e-VLBI, but recognizing the synergy with other geodesy/astronomy
applications requiring real-time or near-real-time high-speed data transmissions. Among the topics to be discussed:
•
•
•
•
•
•
International networking facilities – now and future
User requirements for high-speed networking
Reports on current and projected e-VLBI and related efforts
Networking protocols for real-time data transmission
Public vs. dedicated networks
Establishing international standards for e-VLBI data transfer
Important Dates
•
•
•
•
•
•
3 December 2001- First announcement
1 February 2002 – Second announcement
24 February 2002 – Deadline for abstract submission
1 March 2002 – Final announcement
10 March 2002 – Deadline for registration
22 March 2002 – Deadline for hotel reservations
Program Committee
•
•
•
•
Alan Whitney – Haystack (USA)
Yasuhiro Koyama – CRL (Japan)
Steve Parsley – JIVE (Europe)
Jon Romney – NRAO (USA)
Registration is required for attendance
Registration closes 10 March 2002
Registration is limited to 80, so register early!
For complete information see the meeting web site at:
http://web.haystack.mit.edu/e-vlbi/meeting.html
e-VLBI Development at Haystack Observatory
•
Phase 1: Develop eVLBI-compatible data system
– Mark 5 system
•
Phase 2: Demonstrate 1 Gbps e-VLBI using Bossnet
(w/ DARPA and NASA support)
– ~700km link between Haystack Observatory and NASA/GSFC
– First e-VLBI experiment achieved ~788Mbps transfer rate
•
Phase 3: Develop adaptive network protocol
(newly awarded NSF grant to Haystack Observatory; collaboration with MIT Lab for Computer
Science and MIT Lincoln Laboratory);
– New IP-based protocol tailored to operate in shared-network ‘background’ to
efficiently using available bandwidth
– Demonstrate on national and international networks
•
Phase 4: Extend e-VLBI to national and global VLBI community
–
‘Last mile’ problem remains a serious challenge
Mark 5 VLBI Disk-Based Data System (Phase 1)
Bossnet 1 Gbps e-VLBI
demonstration experiment
(Phase 2)
Westford
Haystack
(correlator)
Future
Initial experiment
USNO
(correlator)
NASA/GSFC
Performance test results – Haystack/GGAO
Average sustained rate >900 Mbps over 10 hours
Correlation results of e-VLBI test observation
Mk4 Fringe Plot
4C39_25.pufxhs, 277-1815, EZ
WESTFORD - GGAO7108, fgroup X, pol RR
Fringe quality
9
SNR
34.5
PFD
0.0e+00
Intg.time 590.412
Amp
2.373
Phase
-52.2
Sbdelay (us)
-0.007122
Mbdelay (us)
0.000000
Fr. rate (Hz)
0.005924
Ref freq (MHz)
8932.9900
AP (sec) 1.000
Exp.
Exper #
2992
Yr:day 2002:277
Start 181500.00
Stop 182500.00
FRT 182000.00
Corr. date:
2002:279:010613
Fourfit date:
2003:006:213921
Position (J2000)
09h27m 3.0139s
+39Ê02'20.852"
International e-VLBI experiments
• Westford, MA to Kashima, Japan - experiments in Oct 02 and Mar 03
– Files exchanged over Abilene/GEMnet networks
• Nominal speed expected to be ~20 Mbps; best achieved so far ~11 Mbps
– Correlation on Mark 4 correlator at Haystack and PC Software correlator
at Kashima; nominal fringes obtained
– Further experiments are scheduled; network tuning is in progress
• Kauai, Hawaii to Wettzell, Germany (in progress)
– Daily experiments of ~100GB are ideal candidate for early e-VLBI
– Data will be transferred to Haystack Observatory for processing
(OC-3 speeds are possible)
– Network links are now being brought up
New IP Protocols for e-VLBI (Phase 3)
• Based on observed usage statistics of networks such as Abilene, it is
clear there is much unused capacity
• New protocols are being developed to utilize networks in ‘background’
mode for applications such as e-VLBI
–
–
–
–
Will ‘scavenge’ and use ‘secondary’ bandwidth
Will give priority to ‘normal’ users
Requires a new ‘end-point adaptive strategy’
Should substantially reduce cost of e-VLBI fiber bandwidth
• Work being carried out under NSF sponsorship by MIT Haystack
Observatory in collaboration with MIT Laboratory for Computer
Science and MIT Lincoln Laboratory
– 3-year program; will demonstrate e-VLBI connections both nationally and
internationally
Typical bit-rate statistics on Abilene network
1.0
Usage >20Mbps less than 1% of the time
0.1
0.01
0.001
100
500
Mbps
Conclusion: Average network usage is only a few % of capacity
Typical distribution of heavy traffic on Abilene
1.0
0.9
0.8
0.7
<10% of ‘bulk’ transfers exceed ~100 secs
200
400
1000
secs
Conclusion: Heavy usage of network tends to occur in bursts of <2 minutes
Extend to national and global community (Phase 4)
• Many possibilities for international connections
–
–
–
–
–
–
Surfnet – U.S. to Europe at 2.5 Gbps
IEEAF – U.S. to Europe at 10 Gbps
TransPAC – U.S. to Japan at 655 Mbps (upgrade to 1.2 Gbps planned)
GEMnet – currently ~20 Mbps, planning to upgrade to 2.2 Gbps
Super-Sinet – 2.5 Gbps Japan-to-U.S.
AMPATH – possible connections to telescopes in Chile and Brazil
• Not well covered
–
–
–
–
Australia
Africa
China
Large parts of South America
But there is a significant problem – ‘
Last-mile’ costs
• Most of the world’s telescopes are not well connected
• Electronic and electro-optic costs are dropping rapidly
– GigE switch:
2001 - $15K 2003 - $1.2K
– GigE transceivers 2001 - $750 2003 - $180
– CWDM transceivers $400-800 for 50-100km reach!
• Direct fiber cost is relatively low– $60/fiber-km in 80-fiber bundle
• If you can buy or lease existing fiber, there is no better time!
• But – fiber installation cost is still tall pole
– Europe: >$20/m (or any populous wide-area)
– U.S.: >$10m (in simplest desert environment)
• The upside: there is developing a lot of momentum and support from
the greater networking community to get the job done!
Also desperately needed:
– Modern digital filter banks to replace aging and obsolete analog BBC’s!
10 Gbps
GÉANT:
The connectivity at 10 Gbps
10 Gbps
2.5 Gbps
EE
LV
34
155
UK
155
LU
NL
155
155
BE
FR
SE
IE
LT
45
SE - PoP for Nordunet
155
DE
ES
PT
CZ
0
SK
622
CH
IT
AT
HU
622
34
HR
BG
622
GR
34
45
CY
155
SI
34
622
PL
IL
34
RO
622 Mbps +10 Gbps l
Transoceanic donations to IEEAF (in red)
Conclusions
• Modern PC and networking technology has led to a revolution in
VLBI data technology
• Transition to all disks will be rapid due to technical, economic and
practical advantages
• e-VLBI is riding an unprecedented wave of global network
connectivity and networking community enthusiasm
• There is no better time to lease or buy installed fiber than now!
• Gradual transition from disks to disk/e-VLBI to all e-VLBI is likely
• 10-100 Gbps/antenna is technically possible with e-VLBI; can VLBI
correlators keep up?
But big hurdles remain…..
• Can e-VLBI survive the long-term networking costs?
• There is momentum gathering in the networking community to provide
national and international ultra-high-speed networking as a critical
‘enabling infrastructure’ for U.S. and international science; the astronomy
community needs to makes its voice heard loud and clear!
EVN Future Vision