SC|05 Bandwidth Challenge ESCC Meeting 9th February ‘06 Yee-Ting Li Stanford Linear Accelerator Center ESCC Meeting 9th February ‘06 Yee-Ting Li Stanford Linear Accelerator Center
LHC Network Requirements CERN/Outside Resource Ratio ~1:2 Tier0/( Tier1)/( Tier2) ~1:1:1 Tier 1 Tier2 Center Online System CERN Center PBs of Disk; Tape Robot FNAL Center IN2P3 Center INFN Center RAL Center Institute Workstations ~ MBytes/sec ~10 Gbps 1 to 10 Gbps Tens of Petabytes by An Exabyte ~5-7 Years later. Physics data cache ~PByte/sec Gbps Tier2 Center ~1-10 Gbps Tier 0 +1 Tier 3 Tier 4 Tier2 Center Tier 2 Experiment
Overview Bandwidth Challenge ‘ The Bandwidth Challenge highlights the best and brightest in new techniques for creating and utilizing vast rivers of data that can be carried across advanced networks. ‘ Transfer as much data as possible using real applications over a 2 hour window We did… Distributed TeraByte Particle Physics Data Sample Analysis ‘Demonstrated high speed transfers of particle physics data between host labs and collaborating institutes in the USA and worldwide. Using state of the art WAN infrastructure and Grid Web Services based on the LHC Tiered Architecture, they showed real-time particle event analysis requiring transfers of Terabyte-scale datasets.’ Bandwidth Challenge ‘ The Bandwidth Challenge highlights the best and brightest in new techniques for creating and utilizing vast rivers of data that can be carried across advanced networks. ‘ Transfer as much data as possible using real applications over a 2 hour window We did… Distributed TeraByte Particle Physics Data Sample Analysis ‘Demonstrated high speed transfers of particle physics data between host labs and collaborating institutes in the USA and worldwide. Using state of the art WAN infrastructure and Grid Web Services based on the LHC Tiered Architecture, they showed real-time particle event analysis requiring transfers of Terabyte-scale datasets.’
Overview In detail, during the bandwidth challenge (2 hours): 131 Gbps measured by SCInet BWC team on 17 of our waves (15 minute average) 95.37TB of data transferred. (3.8 DVD’s per second) Gbps (peak 150.7Gbps) On day of challenge Transferred ~475TB ‘practising’ (waves were shared, still tuning applications and hardware) Peak one way USN utlisation observed on a single link was 9.1Gbps (Caltech) and 8.4Gbps (SLAC) Also wrote to StorCloud SLAC: wrote 3.2TB in 1649 files during BWC Caltech: 6GB/sec with 20 nodes In detail, during the bandwidth challenge (2 hours): 131 Gbps measured by SCInet BWC team on 17 of our waves (15 minute average) 95.37TB of data transferred. (3.8 DVD’s per second) Gbps (peak 150.7Gbps) On day of challenge Transferred ~475TB ‘practising’ (waves were shared, still tuning applications and hardware) Peak one way USN utlisation observed on a single link was 9.1Gbps (Caltech) and 8.4Gbps (SLAC) Also wrote to StorCloud SLAC: wrote 3.2TB in 1649 files during BWC Caltech: 6GB/sec with 20 nodes
Participants Caltech/HEP/CACR/ NetLab: Harvey Newman, Julian Bunn - Contact, Dan Nae, Sylvain Ravot, Conrad Steenberg, Yang Xia, Michael Thomas Caltech SLAC/IEPM: Les Cottrell, Gary Buhrmaster, Yee-Ting Li, Connie Logg SLAC FNAL Matt Crawford, Don Petravick, Vyto Grigaliunas, Dan Yocum FNAL University of Michigan Shawn McKee, Andy Adamson, Roy Hockett, Bob Ball, Richard French, Dean Hildebrand, Erik Hofer, David Lee, Ali Lotia, Ted Hanss, Scott Gerstenberger University of Michigan Caltech/HEP/CACR/ NetLab: Harvey Newman, Julian Bunn - Contact, Dan Nae, Sylvain Ravot, Conrad Steenberg, Yang Xia, Michael Thomas Caltech SLAC/IEPM: Les Cottrell, Gary Buhrmaster, Yee-Ting Li, Connie Logg SLAC FNAL Matt Crawford, Don Petravick, Vyto Grigaliunas, Dan Yocum FNAL University of Michigan Shawn McKee, Andy Adamson, Roy Hockett, Bob Ball, Richard French, Dean Hildebrand, Erik Hofer, David Lee, Ali Lotia, Ted Hanss, Scott Gerstenberger University of Michigan U Florida Paul Avery, Dimitri Bourilkov, U Florida University of Manchester: Richard Hughes-Jones ・ University of Manchester CERN, Switzerland David Foster CERN, Switzerland KAIST, Korea Yusung Kim, KAIST, Korea Kyungpook Univserity, Korea, Kihwan Kwon, Kyungpook Univserity, Korea UERJ, Brazil Alberto Santoro, UERJ, Brazil UNESP, Brazil Sergio Novaes, UNESP, Brazil USP, Brazil Luis Fernandez Lopez USP, Brazil GLORIAD, USA: Greg Cole, Natasha Bulashova GLORIAD U Florida Paul Avery, Dimitri Bourilkov, U Florida University of Manchester: Richard Hughes-Jones ・ University of Manchester CERN, Switzerland David Foster CERN, Switzerland KAIST, Korea Yusung Kim, KAIST, Korea Kyungpook Univserity, Korea, Kihwan Kwon, Kyungpook Univserity, Korea UERJ, Brazil Alberto Santoro, UERJ, Brazil UNESP, Brazil Sergio Novaes, UNESP, Brazil USP, Brazil Luis Fernandez Lopez USP, Brazil GLORIAD, USA: Greg Cole, Natasha Bulashova GLORIAD
Networking Overview We had 22 10Gbits/s waves to the Caltech and SLAC/FNAL booths. Of these: 15 waves to the Caltech booth (from Florida (1), Korea/GLORIAD (1), Brazil (1 * 2.5Gbits/s), Caltech (2), LA (2), UCSD, CERN (2), U Michigan (3), FNAL(2)). 7 x 10Gbits/s waves to the SLAC/FNAL booth (2 from SLAC, 1 from the UK, and 4 from FNAL). The waves were provided by Abilene, Canarie, Cisco (5), ESnet (3), GLORIAD (1), HOPI (1), Michigan Light Rail (MiLR), National Lambda Rail (NLR), TeraGrid (3) and UltraScienceNet (4). We had 22 10Gbits/s waves to the Caltech and SLAC/FNAL booths. Of these: 15 waves to the Caltech booth (from Florida (1), Korea/GLORIAD (1), Brazil (1 * 2.5Gbits/s), Caltech (2), LA (2), UCSD, CERN (2), U Michigan (3), FNAL(2)). 7 x 10Gbits/s waves to the SLAC/FNAL booth (2 from SLAC, 1 from the UK, and 4 from FNAL). The waves were provided by Abilene, Canarie, Cisco (5), ESnet (3), GLORIAD (1), HOPI (1), Michigan Light Rail (MiLR), National Lambda Rail (NLR), TeraGrid (3) and UltraScienceNet (4).
Network Overview
Hardware (SLAC only) At SLAC: 14 x 1.8Ghz Sun v20z (Dual Opteron) 2 x Sun 3500 Disk trays (2TB of storage) 12 x Chelsio T110 10Gb NICs (LR) 2 x Neterion/S2io Xframe I (SR) Dedicated Cisco 6509 with 4 x 4x10GB blades At SC|05: 14 x 2.6Ghz Sun v20z (Dual Opteron) 10 QLogic HBA’s for StorCloud Access 50TB Storage at SC|05 provide by 3PAR (Shared with Caltech) 12 x Neterion/S2io Xframe I NICs (SR) 2 x Chelsio T110 NICs (LR) Shared Cisco 6509 with 6 x 4x10GB blades At SLAC: 14 x 1.8Ghz Sun v20z (Dual Opteron) 2 x Sun 3500 Disk trays (2TB of storage) 12 x Chelsio T110 10Gb NICs (LR) 2 x Neterion/S2io Xframe I (SR) Dedicated Cisco 6509 with 4 x 4x10GB blades At SC|05: 14 x 2.6Ghz Sun v20z (Dual Opteron) 10 QLogic HBA’s for StorCloud Access 50TB Storage at SC|05 provide by 3PAR (Shared with Caltech) 12 x Neterion/S2io Xframe I NICs (SR) 2 x Chelsio T110 NICs (LR) Shared Cisco 6509 with 6 x 4x10GB blades
Hardware at SC|05
Software BBCP ‘Babar File Copy’ Uses ‘ssh’ for authentication Multiple stream capable Features ‘rate synchronisation’ to reduce byte retransmissions Sustained over 9Gbps on a single session XrootD Library for transparent file access (standard unix file functions) Designed primarily for LAN access (transaction based protocol) Managed over 35Gbit/sec (in two directions) on 2 x 10Gbps waves Transferred 18TBytes in 257,913 files DCache 20Gbps production and test cluster traffic BBCP ‘Babar File Copy’ Uses ‘ssh’ for authentication Multiple stream capable Features ‘rate synchronisation’ to reduce byte retransmissions Sustained over 9Gbps on a single session XrootD Library for transparent file access (standard unix file functions) Designed primarily for LAN access (transaction based protocol) Managed over 35Gbit/sec (in two directions) on 2 x 10Gbps waves Transferred 18TBytes in 257,913 files DCache 20Gbps production and test cluster traffic
Last year (SC|04) BWC Aggregate Bandwidth
Cumulative Data Transferred Bandwidth Challenge period
Component Traffic
SLAC-ESnet FermiLab-HOPI SLAC-ESnet-USNFNAL-UltraLight UKLight Out from booth SLAC-FermiLab-UK Bandwidth Contributions In to booth
In to booth Out from booth ESnet routed ESnet SDN layer 2 via USN Bandwidth Challenge period SLAC Cluster Contributions
SLAC/FNAL Booth Aggregate Mbps Waves
Problems… Managerial/PR Initial request for loan hardware took place 6 months in advance! Lots and lots of paperwork to keep account of all loan equipment Logistical Set up and tore down a pseudo production network and servers in a space of week! Testing could not begin until waves were alight Most waves lit day before challenge! Shipping so much hardware not cheap! Setting up monitoring Managerial/PR Initial request for loan hardware took place 6 months in advance! Lots and lots of paperwork to keep account of all loan equipment Logistical Set up and tore down a pseudo production network and servers in a space of week! Testing could not begin until waves were alight Most waves lit day before challenge! Shipping so much hardware not cheap! Setting up monitoring
Problems… Tried to configure hardware and software prior to show Hardware NICS We had 3 bad Chelsios (bad memory) Xframe II’s did not work in UKLight’s Boston machines Hard-disks 3 dead 10K disks (had to ship in spare) 1 x 4Port 10Gb blade DOA MTU mismatch between domains Router blade died during stress testing day before BWC! Cables! Cables! Cables! Software Used golden disks for duplication (still takes 30 minutes per disk to replicate!) Linux kernels: Initially used , found sever performance problems compared to (New) Router firmware caused crashes under heavy load Unfortunately, only discovered just before BWC Had to manually restart the affected ports during BWC Tried to configure hardware and software prior to show Hardware NICS We had 3 bad Chelsios (bad memory) Xframe II’s did not work in UKLight’s Boston machines Hard-disks 3 dead 10K disks (had to ship in spare) 1 x 4Port 10Gb blade DOA MTU mismatch between domains Router blade died during stress testing day before BWC! Cables! Cables! Cables! Software Used golden disks for duplication (still takes 30 minutes per disk to replicate!) Linux kernels: Initially used , found sever performance problems compared to (New) Router firmware caused crashes under heavy load Unfortunately, only discovered just before BWC Had to manually restart the affected ports during BWC
Problems Most transfers were from memory to memory (Ramdisk etc). Local caching of (small) files in memory Reading and writing to disk will be the next bottleneck to overcome Most transfers were from memory to memory (Ramdisk etc). Local caching of (small) files in memory Reading and writing to disk will be the next bottleneck to overcome
Conclusion Previewed the IT Challenges of the next generation Data Intensive Science Applications (High Energy Physics, astronomy etc) Petabyte-scale datasets Tens of national and transoceanic links at 10 Gbps (and up) 100+ Gbps aggregate data transport sustained for hours; We reached a Petabyte/day transport rate for real physics data Learned to gauge difficulty of the global networks and transport systems required for the LHC mission Set up, shook down and successfully ran the systems in < 1 week Understood and optimized the configurations of various components (Network interfaces, router/switches, OS, TCP kernels, applications) for high performance over the wide area network. Previewed the IT Challenges of the next generation Data Intensive Science Applications (High Energy Physics, astronomy etc) Petabyte-scale datasets Tens of national and transoceanic links at 10 Gbps (and up) 100+ Gbps aggregate data transport sustained for hours; We reached a Petabyte/day transport rate for real physics data Learned to gauge difficulty of the global networks and transport systems required for the LHC mission Set up, shook down and successfully ran the systems in < 1 week Understood and optimized the configurations of various components (Network interfaces, router/switches, OS, TCP kernels, applications) for high performance over the wide area network.
Conclusion Products from this the exercise An optimized Linux ( NFSv4 + FAST and other TCP stacks) kernel for data transport; after 7 full kernel-build cycles in 4 days A newly optimized application-level copy program, bbcp, that matches the performance of iperf under some conditions. Extensions of Xrootd, an optimized low-latency file access application for clusters, across the wide area Understanding of the limits of 10 Gbps-capable systems under stress. How to effectively utilize 10GE and 1GE connected systems to drive 10 gigabit wavelengths in both directions. Use of production and test clusters at FNAL reaching more than 20 Gbps of network throughput. Significant efforts remain from the perspective of high-energy physics Management, integration and optimization of network resources End-to-end capabilities able to utilize these network resources. This includes applications and IO devices (disk and storage systems) Products from this the exercise An optimized Linux ( NFSv4 + FAST and other TCP stacks) kernel for data transport; after 7 full kernel-build cycles in 4 days A newly optimized application-level copy program, bbcp, that matches the performance of iperf under some conditions. Extensions of Xrootd, an optimized low-latency file access application for clusters, across the wide area Understanding of the limits of 10 Gbps-capable systems under stress. How to effectively utilize 10GE and 1GE connected systems to drive 10 gigabit wavelengths in both directions. Use of production and test clusters at FNAL reaching more than 20 Gbps of network throughput. Significant efforts remain from the perspective of high-energy physics Management, integration and optimization of network resources End-to-end capabilities able to utilize these network resources. This includes applications and IO devices (disk and storage systems)
Press and PR 11/8/05 - Brit Boffins aim to Beat LAN speed record from vnunet.comBrit Boffins aim to Beat LAN speed record SC|05 Bandwidth Challenge SLAC Interaction Point. SC|05 Bandwidth Challenge Top Researchers, Projects in High Performance Computing Honored at SC/05... Business Wire (press release) - San Francisco, CA, USA Top Researchers, Projects in High Performance Computing Honored at SC/05... 11/18/05 - Official Winner AnnouncementOfficial Winner Announcement 11/18/05 - SC|05 Bandwidth Challenge Slide PresentationSC|05 Bandwidth Challenge Slide Presentation 11/23/05 - Bandwidth Challenge Results from SlashdotBandwidth Challenge Results 12/6/05 - Caltech press releaseCaltech press release 12/6/05 - Neterion Enables High Energy Physics Team to Beat World Record Speed at SC05 Conference CCN Matthews News Distribution ExpertsNeterion Enables High Energy Physics Team to Beat World Record Speed at SC05 Conference High energy physics team captures network prize at SC|05 from SLAC High energy physics team captures network prize at SC|05 High energy physics team captures network prize at SC|05 EurekaAlert! High energy physics team captures network prize at SC|05 12/7/05 - High Energy Physics Team Smashes Network Record, from Science Grid this Week.High Energy Physics Team Smashes Network Record Congratulations to our Research Partners for a New Bandwidth Record at SuperComputing 2005, from Neterion. Congratulations to our Research Partners for a New Bandwidth Record at SuperComputing 2005 11/8/05 - Brit Boffins aim to Beat LAN speed record from vnunet.comBrit Boffins aim to Beat LAN speed record SC|05 Bandwidth Challenge SLAC Interaction Point. SC|05 Bandwidth Challenge Top Researchers, Projects in High Performance Computing Honored at SC/05... Business Wire (press release) - San Francisco, CA, USA Top Researchers, Projects in High Performance Computing Honored at SC/05... 11/18/05 - Official Winner AnnouncementOfficial Winner Announcement 11/18/05 - SC|05 Bandwidth Challenge Slide PresentationSC|05 Bandwidth Challenge Slide Presentation 11/23/05 - Bandwidth Challenge Results from SlashdotBandwidth Challenge Results 12/6/05 - Caltech press releaseCaltech press release 12/6/05 - Neterion Enables High Energy Physics Team to Beat World Record Speed at SC05 Conference CCN Matthews News Distribution ExpertsNeterion Enables High Energy Physics Team to Beat World Record Speed at SC05 Conference High energy physics team captures network prize at SC|05 from SLAC High energy physics team captures network prize at SC|05 High energy physics team captures network prize at SC|05 EurekaAlert! High energy physics team captures network prize at SC|05 12/7/05 - High Energy Physics Team Smashes Network Record, from Science Grid this Week.High Energy Physics Team Smashes Network Record Congratulations to our Research Partners for a New Bandwidth Record at SuperComputing 2005, from Neterion. Congratulations to our Research Partners for a New Bandwidth Record at SuperComputing 2005
SLAC/UK Contribution ESnet/USN layer 2 UKLight In to booth Out from booth ESnet routed
SLAC/Esnet Contribution Mbps Hosts Aggregate
HOPI USN FermiLab Contribution UltraLight