Steven Carter. Network Lead, NCCS Oak Ridge National Laboratory OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 1
|
|
- Noah Sherman
- 5 years ago
- Views:
Transcription
1 Networking the National Leadership Computing Facility Steven Carter Network Lead, NCCS Oak Ridge National Laboratory 1
2 Outline Introduction NCCS Network Infrastructure Cray Architecture Overview NCCS Enhancements Future Work Summary 2
3 Introduction Big Machine. Needs to be shared. The goal is to enable science: Area Present Beyond Remarks High Energy Physics 100 Gb/s 1 Tb/s High Throughput Climate Gb/s n Tb/s High Throughput SNS Nanoscience 1 Gb/s n Tb/s Remote Control & High Throughput Most of our work has been specifically aimed at make applications more productive (e.g. TSI). 3
4 Introduction (Cont.) Many problems trying to get to 10G: Stock TCP has problems at high speeds. Many possible solutions: TCP variants, UDP protocols, L2 protocols, etc. 10G cards are a burden on the system. This is getting better: TOE, less memory movement. Many system components cannot handle 10G rates (e.g. PCI-X too slow). PCI-Express on the way out. Many disk subsystems cannot handle 10G rates. Still a big problem. Security (Firewalls, IDS) has not caught up. Some 10G solutions: MeteNetworks, Endace. 4
5 NCCS Network Infrastructure (LAN) 10G to WAN The NCCS is making a substantial commitment in localand wide-area networking to enable the NLCF machines to produce good science. Local-Area Network: 3 x Cisco 6500 series switches. 10G aggregation switch. 10G Backbone. Hybrid firewall/acls. 10G interface to wide-area. Try to reach a good compromise between security and performance. HPSS Ram (Altix) Cheetah (Power4) External Enclaves Cisco 6513 FWSM MFSC Cisco 6509 MFSC MFSC 10G Backbone Cisco 6509 FWSM MFSC Tiger (XD1) Phoenix (X1) Jaguar (XT3) Internal Enclaves Administrative Management Services 5
6 NCCS Network Infrastructure (WAN) Wide-Area Network: ORNL Connector: 1Tb/s Total Capacity Will provide 10G circuits to ESNet, Internet 2, Teragrid, UltraScience Net, Cheetah Network. DOE UltraScience Net, NSF Cheetah Net: Developing technology to enable application controlled dedicated circuits (among other things). UltraScience and Cheetah will peer at ORNL giving coast to coast access. UltraScience Net Ciena CoreDirector NCCS Network FutureNet Ciena CoreStream TeraGrid Juniper T320 Campus Juniper T640 OC G WANPhy 10G LANPhy 6
7 DOE UltraScience + NSF Cheetah CERN PNNL NERSC CUNY SLAC UVa LBL ORNL CalTech FNAL ANL NCSU University Laboratory Cheetah Network UltraScience Net Future Connection Current Connection 7
8 Cray Architecture Overview Each machine has slightly different ways of performing network I/O. Some are better than others, but you have to know what you are dealing with to use it effectively. 8
9 Cray XD1 Nodes 5 & 6 own the physical interfaces. Nodes 5 & 6 present the MAC & IP addresses for the nodes with virtual interfaces. Bridging is done over the RapidArray. Penalty for using off-node interfaces. Some overhead on nodes 5 & 6. Rapid Routers Array PCI-X Expansion Card GbE LAN 9
10 Cray XT3 Login nodes have 1G interfaces. Network nodes have 10G interfaces. Application nodes cannot open sockets. IP-over-Portals network amongst the nodes. Penalty for using off-node interfaces. Not obvious how to use all of the interfaces effectively. C I N L C C C C C C C C C C C C I I Compute Node I/O Node Netowork Node Login Node N N L L 10GbE LAN 1GbE 10
11 Cray X1 A A A A Socket calls made on App Nodes are suspended and migrated to OS Nodes. ~500 system calls/sec from App Node. ~5000 system calls/sec from OS Node. OS Nodes connect to I/O drawers via SPC. I/O drawers have PCI-X bridges with Fiber Channel cards. CNS connects via Fiber Channel. Routers A A O O 1.6GB/s SPC CNS I/O Drawers 1GbE 2Gb/s FC LAN 11
12 Cray X1: CNS Communicates with X1 via IPover-FC w/ ~64k frames. Iptables diverts TCP streams to tcp_assistd and masquerades incoming and outgoing packets. tcp_assist terminates socket and opens another to the destination. Reading from one socket and writing to the other, tcp_assist either fragments or coalesces packets to match the MTU. tcp_assist forwards TCP packets faster then ip_forward. UDP packet are handled my the stock ip_forward functionality of the system. To X1 I P T ab l e s TCP_assist (conversion each direction) IP Forward (fragmentation on transmit) TCP UDP I P T ab l e s To LAN 12
13 CNS Testing, Round 1 Hardware: Standard CNS w/2 x Emulex 9802DC Software: CNS 1.2 base 2.4 kernel. UDP Knee Net100 modifications: Re-wrote tcp_assistd (ships in CNS 1.4 and above sorry for any problems). WAN performance increase by 400%. Bonded interface testing: 1-4 Interfaces. Proprietary and Open source Emulex drivers tested. X1 I/O Drawer 4 x 1.2GBps SPC Channels 2 x Emulex 9802DC 2Gbps FC HBA Custom CNS 13
14 Round 1 Results Big difference between Open Source and Proprietary Drivers. Proprietary Drivers: Higher throughput. Higher system utilization. Open Source Drivers: Lower throughput. Lower system utilization. Conclusion: System utilization too high to use effectively without 10G NIC with good offload. 14
15 CNS Testing, Round 2 Hardware: 2.2GHz Dual Opteron 2 x Emulex 9802DCs 1 x Chelsio T110 10G NIC. Software: Suse 9.2 base kernel. ORNL s tcp_assistd. Emulex 2.10 drivers. Tested both bonded interfaces and nonbonded interfaces. X1 I/O Drawer 4 x 1.2GBps SPC Channels 2 x Emulex 9802DC 2Gbps FC HBA Chelsio T110 10Gbps NIC Custom CNS Host 15
16 Round 2 Results Throughput of 5Gb/s (10Gb/s aggregate). 1.8 Gb/s file transfer using bbcp over bonded channels. Even unbonded channels can be exploited for single transfers (e.g. GridFTP). 16
17 Future Work Put the SuperCNS into production with the test networks (i.e. UltraScience Net, Cheetah) for TSI. Port IP-over-FC Functionality to Emulex s latest open source drivers (v8). Same type of work with the XD1, and XT3. 17
18 Summary It will be challenging to meet the networking needs stated by the various science areas. A holistic approach needs to be taken achieve these goals (i.e. local- and wide-area, host tuning/design, application modifications). NCCS s current work has paid dividends in enabling scientists to do their work (TSI can now transfer files in hours instead of days).. 18
19 References [Net100] Net [X1Overview] Cray X1 System Overview Version 2.4. [XD1Overview] Cray XD1 System Overview Release 1.1. [XT3Overview] Cray XT3 System Overview Version
20 Acknowledgments This work was sponsored by the U.S. Department of Energy s Office of Advanced Scientific Computing Research and performed at the Oak Ridge National Laboratory, managed by UT-Battelle, LLC under contract number DE-AC05-00OR This work is partially sponsored by the Laboratory Directed Research and Development Project at ORNL. The submitted manuscript has been authored by a contractor of the U.S. Government under Contract No. DE-AC05-00OR Accordingly, the U.S. Government retains a non-exclusive, royalty-free license to publish or reproduce the published form of the contributions, or allow other to do so, for U.S. Government purposes. 20
21 The End Comments? Questions? Criticisms? 21
UltraScience Net Update: Network Research Experiments
UltraScience Net Update: Network Research Experiments Nagi Rao, Bill Wing, Susan Hicks, Paul Newman, Steven Carter Oak Ridge National Laboratory raons@ornl.gov https://www.csm.ornl.gov/ultranet February
More informationA TRANSPORT PROTOCOL FOR DEDICATED END-TO-END CIRCUITS
A TRANSPORT PROTOCOL FOR DEDICATED END-TO-END CIRCUITS MS Thesis Final Examination Anant P. Mudambi Computer Engineering University of Virginia December 6, 2005 Outline Motivation CHEETAH Background UDP-based
More informationWide-Area InfiniBand RDMA: Experimental Evaluation
Wide-Area InfiniBand RDMA: Experimental Evaluation Nagi Rao, Steve Poole, Paul Newman, Susan Hicks Oak Ridge National Laboratory High-Performance Interconnects Workshop August 31, 2009, New Orleans, LA
More informationZhengyang Liu University of Virginia. Oct 29, 2012
SDCI Net: Collaborative Research: An integrated study of datacenter networking and 100 GigE wide-area networking in support of distributed scientific computing Zhengyang Liu University of Virginia Oct
More informationPhilip C. Roth. Computer Science and Mathematics Division Oak Ridge National Laboratory
Philip C. Roth Computer Science and Mathematics Division Oak Ridge National Laboratory A Tree-Based Overlay Network (TBON) like MRNet provides scalable infrastructure for tools and applications MRNet's
More informationAnalytics of Wide-Area Lustre Throughput Using LNet Routers
Analytics of Wide-Area Throughput Using LNet Routers Nagi Rao, Neena Imam, Jesse Hanley, Sarp Oral Oak Ridge National Laboratory User Group Conference LUG 2018 April 24-26, 2018 Argonne National Laboratory
More informationTraffic engineering and GridFTP log analysis. Jan 17, 2013 Project web site:
Traffic engineering and GridFTP log analysis Zhenzhen Yan, Z. Liu, M. Veeraraghavan University of Virginia mvee@virginia.edu Chris Tracy, Chin Guok ESnet ctracy@es.net Jan 17, 2013 Project web site: http://www.ece.virginia.edu/mv/research/doe09/index.html
More informationIndiana University s Lustre WAN: The TeraGrid and Beyond
Indiana University s Lustre WAN: The TeraGrid and Beyond Stephen C. Simms Manager, Data Capacitor Project TeraGrid Site Lead, Indiana University ssimms@indiana.edu Lustre User Group Meeting April 17, 2009
More informationEarly Evaluation of the Cray X1 at Oak Ridge National Laboratory
Early Evaluation of the Cray X1 at Oak Ridge National Laboratory Patrick H. Worley Thomas H. Dunigan, Jr. Oak Ridge National Laboratory 45th Cray User Group Conference May 13, 2003 Hyatt on Capital Square
More informationEngagement With Scientific Facilities
Engagement With Scientific Facilities Eli Dart, Network Engineer ESnet Science Engagement Lawrence Berkeley National Laboratory Global Science Engagement Panel Internet2 Technology Exchange San Francisco,
More informationComparison of Scheduling Policies and Workloads on the NCCS and NICS XT4 Systems at Oak Ridge National Laboratory
Comparison of Scheduling Policies and Workloads on the NCCS and NICS XT4 Systems at Oak Ridge National Laboratory Troy Baer HPC System Administrator National Institute for Computational Sciences University
More informationAchieving 98Gbps of Crosscountry TCP traffic using 2.5 hosts, 10 x 10G NICs, and 10 TCP streams
Achieving 98Gbps of Crosscountry TCP traffic using 2.5 hosts, 10 x 10G NICs, and 10 TCP streams Eric Pouyoul, Brian Tierney ESnet January 25, 2012 ANI 100G Testbed ANI Middleware Testbed NERSC To ESnet
More informationTitle: Collaborative research: End-to-End Provisioned Optical Network Testbed for Large-Scale escience Applications
Year 1 Activities report for the NSF project EIN-0335190 Title: Collaborative research: End-to-End Provisioned Optical Network Testbed for Large-Scale escience Applications Date: July 29, 2004 (this is
More informationGlobus Online and HPSS. KEK, Tsukuba Japan October 16 20, 2017 Guangwei Che
Globus Online and HPSS KEK, Tsukuba Japan October 16 20, 2017 Guangwei Che Agenda (1) What is Globus and Globus Online? How Globus Online works? Globus DSI module for HPSS Globus Online setup DSI module
More informationChapter 4. Routers with Tiny Buffers: Experiments. 4.1 Testbed experiments Setup
Chapter 4 Routers with Tiny Buffers: Experiments This chapter describes two sets of experiments with tiny buffers in networks: one in a testbed and the other in a real network over the Internet2 1 backbone.
More informationEarly Operational Experience with the Cray X1 at the Oak Ridge National Laboratory Center for Computational Sciences
Early Operational Experience with the Cray X1 at the Oak Ridge National Laboratory Center for Computational Sciences Buddy Bland, Richard Alexander Steven Carter, Kenneth Matney, Sr. Cray User s Group
More informationScalaIOTrace: Scalable I/O Tracing and Analysis
ScalaIOTrace: Scalable I/O Tracing and Analysis Karthik Vijayakumar 1, Frank Mueller 1, Xiaosong Ma 1,2, Philip C. Roth 2 1 Department of Computer Science, NCSU 2 Computer Science and Mathematics Division,
More informationOak Ridge National Laboratory
Oak Ridge National Laboratory Lustre Scalability Workshop Presented by: Galen M. Shipman Collaborators: David Dillow Sarp Oral Feiyi Wang February 10, 2009 We have increased system performance 300 times
More informationPreparing GPU-Accelerated Applications for the Summit Supercomputer
Preparing GPU-Accelerated Applications for the Summit Supercomputer Fernanda Foertter HPC User Assistance Group Training Lead foertterfs@ornl.gov This research used resources of the Oak Ridge Leadership
More informationSTORAGE CONSOLIDATION WITH IP STORAGE. David Dale, NetApp
STORAGE CONSOLIDATION WITH IP STORAGE David Dale, NetApp SNIA Legal Notice The material contained in this tutorial is copyrighted by the SNIA. Member companies and individuals may use this material in
More informationZhengyang Liu! Oct 25, Supported by NSF Grant OCI
SDCI Net: Collaborative Research: An integrated study of datacenter networking and 100 GigE wide-area networking in support of distributed scientific computing Zhengyang Liu! Oct 25, 2013 Supported by
More informationAllowing Users to Run Services at the OLCF with Kubernetes
Allowing Users to Run Services at the OLCF with Kubernetes Jason Kincl Senior HPC Systems Engineer Ryan Adamson Senior HPC Security Engineer This work was supported by the Oak Ridge Leadership Computing
More informationFile Transfer: Basics and Best Practices. Joon Kim. Ph.D. PICSciE. Research Computing 09/07/2018
File Transfer: Basics and Best Practices Joon Kim. Ph.D. PICSciE Research Computing Workshop @Chemistry 09/07/2018 Our goal today Learn about data transfer basics Pick the right tool for your job Know
More informationWide-Area Networking at SLAC. Warren Matthews and Les Cottrell (SCS Network Group) Presented at SLAC, April
Wide-Area Networking at SLAC Warren Matthews and Les Cottrell (SCS Network Group) Presented at SLAC, April 6 2001. Overview SLAC s Connections to WANs Utilization End-to-end Performance The Future Note:
More informationEnabling a SuperFacility with Software Defined Networking
Enabling a SuperFacility with Software Defined Networking Shane Canon Tina Declerck, Brent Draney, Jason Lee, David Paul, David Skinner May 2017 CUG 2017-1 - SuperFacility - Defined Combining the capabilities
More informationSTORAGE CONSOLIDATION WITH IP STORAGE. David Dale, NetApp
STORAGE CONSOLIDATION WITH IP STORAGE David Dale, NetApp SNIA Legal Notice The material contained in this tutorial is copyrighted by the SNIA. Member companies and individuals may use this material in
More informationThe Common Communication Interface (CCI)
The Common Communication Interface (CCI) Presented by: Galen Shipman Technology Integration Lead Oak Ridge National Laboratory Collaborators: Scott Atchley, George Bosilca, Peter Braam, David Dillow, Patrick
More informationAchieving Record Speed Trans- Atlantic End-to-end TCP Throughput
Achieving Record Speed Trans- Atlantic End-to-end TCP Throughput Les Cottrell SLAC Prepared for the NANOG meeting, Salt Lake City, June 2003 http://www.slac.stanford.edu/grp/scs/net/talk/nanog-jun03.html
More informationGPFS on a Cray XT. Shane Canon Data Systems Group Leader Lawrence Berkeley National Laboratory CUG 2009 Atlanta, GA May 4, 2009
GPFS on a Cray XT Shane Canon Data Systems Group Leader Lawrence Berkeley National Laboratory CUG 2009 Atlanta, GA May 4, 2009 Outline NERSC Global File System GPFS Overview Comparison of Lustre and GPFS
More informationEarly Evaluation of the Cray XD1
Early Evaluation of the Cray XD1 (FPGAs not covered here) Mark R. Fahey Sadaf Alam, Thomas Dunigan, Jeffrey Vetter, Patrick Worley Oak Ridge National Laboratory Cray User Group May 16-19, 2005 Albuquerque,
More informationIntroduction. High Speed LANs. Emergence of High-Speed LANs. Characteristics of High Speed LANS. Text ch. 6, High-Speed Networks and
High Speed LANs 3BA33 David Lewis 2 nd Semester 2006-07 3BA33 D.Lewis 2007 1 Characteristics of High Speed LANS 3BA33 D.Lewis 2007 3 Introduction Fast Ethernet and Gigabit Ethernet Fibre Channel High-speed
More informationInternet data transfer record between CERN and California. Sylvain Ravot (Caltech) Paolo Moroni (CERN)
Internet data transfer record between CERN and California Sylvain Ravot (Caltech) Paolo Moroni (CERN) Summary Internet2 Land Speed Record Contest New LSR DataTAG project and network configuration Establishing
More informationComparison of XT3 and XT4 Scalability
Comparison of XT3 and XT4 Scalability Patrick H. Worley Oak Ridge National Laboratory CUG 2007 May 7-10, 2007 Red Lion Hotel Seattle, WA Acknowledgements Research sponsored by the Climate Change Research
More informationLecture 1 Overview - Data Communications, Data Networks, and the Internet
DATA AND COMPUTER COMMUNICATIONS Lecture 1 Overview - Data Communications, Data Networks, and the Internet Mei Yang Based on Lecture slides by William Stallings 1 OUTLINE Data Communications and Networking
More informationHigh Performance Ethernet for Grid & Cluster Applications. Adam Filby Systems Engineer, EMEA
High Performance Ethernet for Grid & Cluster Applications Adam Filby Systems Engineer, EMEA 1 Agenda Drivers & Applications The Technology Ethernet Everywhere Ethernet as a Cluster interconnect Ethernet
More informationDisclaimer This presentation may contain product features that are currently under development. This overview of new technology represents no commitme
NET1343BU NSX Performance Samuel Kommu #VMworld #NET1343BU Disclaimer This presentation may contain product features that are currently under development. This overview of new technology represents no
More informationData transfer over the wide area network with a large round trip time
Journal of Physics: Conference Series Data transfer over the wide area network with a large round trip time To cite this article: H Matsunaga et al 1 J. Phys.: Conf. Ser. 219 656 Recent citations - A two
More informationCISL Update. 29 April Operations and Services Division
CISL Update Operations and Services CISL HPC Advisory Panel Meeting Anke Kamrath anke@ucar.edu Operations and Services Division Computational and Information Systems Laboratory 1 CHAP Meeting 14 May 2009
More informationHIGH-PERFORMANCE NETWORKING :: USER-LEVEL NETWORKING :: REMOTE DIRECT MEMORY ACCESS
HIGH-PERFORMANCE NETWORKING :: USER-LEVEL NETWORKING :: REMOTE DIRECT MEMORY ACCESS CS6410 Moontae Lee (Nov 20, 2014) Part 1 Overview 00 Background User-level Networking (U-Net) Remote Direct Memory Access
More informationIPv4 addressing, NAT. Computer Networking: A Top Down Approach 6 th edition Jim Kurose, Keith Ross Addison-Wesley.
IPv4 addressing, NAT http://xkcd.com/195/ Computer Networking: A Top Down Approach 6 th edition Jim Kurose, Keith Ross Addison-Wesley Some materials copyright 1996-2012 J.F Kurose and K.W. Ross, All Rights
More information8-Port 10/100TX 802.3at PoE + 2-Port Gigabit TP/SFP Combo Web Smart Ethernet Switch FGSD-1008HPS
8-Port 10/100TX 802.3at PoE + 2-Port Gigabit TP/SFP Combo Web Smart Ethernet Switch FGSD-1008HPS Presentation Outlines Product Overview Product Features Product Benefits Applications 2 / 34 Product Overview
More informationPCIe 10G SFP+ Network Card
PCIe 10G SFP+ Network Card User Manual Ver. 1.00 All brand names and trademarks are properties of their respective owners. Contents: Chapter 1: Introduction... 3 1.1 Product Introduction... 3 1.2 Features...
More informationWide-Area Performance Profiling of 10GigE and InfiniBand Technologies
Wide-Area Performance Profiling of 1GigE and InfiniBand Technologies Nageswara S. V. Rao, Weikuan Yu, William R. Wing, Stephen W. Poole, and Jeffrey S. Vetter Oak Ridge National Laboratory, Oak Ridge,
More informationSLIDE 1 - COPYRIGHT 2015 ELEPHANT FLOWS IN THE ROOM: SCIENCEDMZ NATIONALLY DISTRIBUTED
SLIDE 1 - COPYRIGHT 2015 ELEPHANT FLOWS IN THE ROOM: SCIENCEDMZ NATIONALLY DISTRIBUTED SLIDE 2 - COPYRIGHT 2015 Do you know what your campus network is actually capable of? (i.e. have you addressed your
More informationBigData Express: Toward Predictable, Schedulable, and High-Performance Data Transfer. BigData Express Research Team November 10, 2018
BigData Express: Toward Predictable, Schedulable, and High-Performance Data Transfer BigData Express Research Team November 10, 2018 Many people s hard work FNAL: ESnet: icair/starlight: KISTI: Qiming
More informationData & Computer Communication
Basic Networking Concepts A network is a system of computers and other devices (such as printers and modems) that are connected in such a way that they can exchange data. A bridge is a device that connects
More informationCTCP (Circuit TCP) v1.0
Outline C (Circuit ) v1.0 Helali Bhuiyan and Malathi Veeraraghavan {helali,mv5g}@virginia.edu Purpose of C Requirements to run C code C Components C Operation C details C Usage Usage of C across an Internet
More informationClearStream. Prototyping 40 Gbps Transparent End-to-End Connectivity. Cosmin Dumitru! Ralph Koning! Cees de Laat! and many others (see posters)!
ClearStream Prototyping 40 Gbps Transparent End-to-End Connectivity Cosmin Dumitru! Ralph Koning! Cees de Laat! and many others (see posters)! University of Amsterdam! more data! Speed! Volume! Internet!
More informationHigh Speed Asynchronous Data Transfers on the Cray XT3
High Speed Asynchronous Data Transfers on the Cray XT3 Ciprian Docan, Manish Parashar and Scott Klasky The Applied Software System Laboratory Rutgers, The State University of New Jersey CUG 2007, Seattle,
More informationVCOM Device Interface User Guide
VCOM Device Interface User Guide Patent Pending: U.S. Serial # 11/970,871 Revised: October 2013 IntraCom Systems, LLC Ph: (888) 318-6187 support@intracomsystems.com www.intracomsystems.com Copyright 2013
More informationThe router architecture consists of two major components: Routing Engine. 100-Mbps link. Packet Forwarding Engine
Chapter 4 The router architecture consists of two major components: Packet Forwarding Engine Performs Layer 2 and Layer 3 packet switching, route lookups, and packet forwarding. Routing Engine Provides
More informationProgrammable Information Highway (with no Traffic Jams)
Programmable Information Highway (with no Traffic Jams) Inder Monga Energy Sciences Network Scientific Networking Division Lawrence Berkeley National Lab Exponential Growth ESnet Accepted Traffic: Jan
More informationChapter 2 Communicating Over the Network
Chapter 2 Communicating Over the Network Elements of Communication Communicating the Messages Continuous stream of bits 00101010100101010101010101010101010 I have to wait Single communications (e.g. video,
More informationAddressing and secure control-plane network design in GMPLS networks
Addressing and secure control-plane network design in GMPLS networks Abstract 1 : Malathi Veeraraghavan, Xuan Zheng, Xiangfei Zhu {malathi, xuan, xz4p}@virginia.edu April 7, 2006 This document describes
More informationRapid Platform Deployment: Allows clients to concentrate their efforts on application software.
Overview Aliathon Ltd. in partnership with Nallatech brings to market a demo design based on the Universal Network Probe technology described in Aliathon Application Note 06. This design demonstrate the
More informationHow Cisco IT Deployed Cisco Firewall Services Modules at Scientific Atlanta
How Cisco IT Deployed Cisco Firewall Services Modules at Scientific Atlanta Deployment increases network security, reliability, and availability. Cisco IT Case Study / Security / Firewall Services Module:
More informationScaling a Global File System to the Greatest Possible Extent, Performance, Capacity, and Number of Users
Scaling a Global File System to the Greatest Possible Extent, Performance, Capacity, and Number of Users Phil Andrews, Bryan Banister, Patricia Kovatch, Chris Jordan San Diego Supercomputer Center University
More informationHPC Saudi Jeffrey A. Nichols Associate Laboratory Director Computing and Computational Sciences. Presented to: March 14, 2017
Creating an Exascale Ecosystem for Science Presented to: HPC Saudi 2017 Jeffrey A. Nichols Associate Laboratory Director Computing and Computational Sciences March 14, 2017 ORNL is managed by UT-Battelle
More informationRouters: Forwarding EECS 122: Lecture 13
Routers: Forwarding EECS 122: Lecture 13 epartment of Electrical Engineering and Computer Sciences University of California Berkeley Router Architecture Overview Two key router functions: run routing algorithms/protocol
More informationIntroduction to FREE National Resources for Scientific Computing. Dana Brunson. Jeff Pummill
Introduction to FREE National Resources for Scientific Computing Dana Brunson Oklahoma State University High Performance Computing Center Jeff Pummill University of Arkansas High Peformance Computing Center
More informationCisco Wide Area Application Services and Cisco Nexus Family Switches: Enable the Intelligent Data Center
Cisco Wide Area Application Services and Cisco Nexus Family Switches: Enable the Intelligent Data Center What You Will Learn IT departments are facing increasing pressure to accommodate numerous changing
More informationRouters: Forwarding EECS 122: Lecture 13
Input Port Functions Routers: Forwarding EECS 22: Lecture 3 epartment of Electrical Engineering and Computer Sciences University of California Berkeley Physical layer: bit-level reception ata link layer:
More informationRouter Router Microprocessor controlled traffic direction home router DSL modem Computer Enterprise routers Core routers
Router Router is a Microprocessor controlled device that forwards data packets across the computer network. It is used to connect two or more data lines from different net works. The function of the router
More informationSolaris Engineered Systems
Solaris Engineered Systems SPARC SuperCluster Introduction Andy Harrison andy.harrison@oracle.com Engineered Systems, Revenue Product Engineering The following is intended to outline
More informationiscsi Technology: A Convergence of Networking and Storage
HP Industry Standard Servers April 2003 iscsi Technology: A Convergence of Networking and Storage technology brief TC030402TB Table of Contents Abstract... 2 Introduction... 2 The Changing Storage Environment...
More informationPath-based networking: From POTS to SDN. Outline
Path-based networking: From POTS to SDN Malathi Veeraraghavan University of Virginia April 28, 2014 Talk at CUHK, Dept. of IE Thanks to the US DOE ASCR for grant DE-SC0007341, and NSF grants CNS-1116081,
More informationIsoStack Highly Efficient Network Processing on Dedicated Cores
IsoStack Highly Efficient Network Processing on Dedicated Cores Leah Shalev Eran Borovik, Julian Satran, Muli Ben-Yehuda Outline Motivation IsoStack architecture Prototype TCP/IP over 10GE on a single
More informationVirtual Dispersive Networking Spread Spectrum IP
Virtual Dispersive Networking Spread Spectrum IP DSI Proprietary 1 DSI Proprietary 2 Problem Lies Outside of Existing Security: On the Internet Internet Routers Virus Software Phishing Software etc POLICY
More informationRonald van der Pol
Ronald van der Pol Contributors! " Ronald van der Pol! " Freek Dijkstra! " Pieter de Boer! " Igor Idziejczak! " Mark Meijerink! " Hanno Pet! " Peter Tavenier (this work is partially funded
More informationModule 2 Storage Network Architecture
Module 2 Storage Network Architecture 1. SCSI 2. FC Protocol Stack 3. SAN:FC SAN 4. IP Storage 5. Infiniband and Virtual Interfaces FIBRE CHANNEL SAN 1. First consider the three FC topologies pointto-point,
More informationJMR ELECTRONICS INC. WHITE PAPER
THE NEED FOR SPEED: USING PCI EXPRESS ATTACHED STORAGE FOREWORD The highest performance, expandable, directly attached storage can be achieved at low cost by moving the server or work station s PCI bus
More informationEnterprise. Nexus 1000V. L2/L3 Fabric WAN/PE. Customer VRF. MPLS Backbone. Service Provider Data Center-1 Customer VRF WAN/PE OTV OTV.
2 CHAPTER Cisco's Disaster Recovery as a Service (DRaaS) architecture supports virtual data centers that consist of a collection of geographically-dispersed data center locations. Since data centers are
More informationBuilding 10-Gbps Networks: A few observations on the national and regional scales in the U.S.
Building 10-Gbps Networks: A few observations on the national and regional scales in the U.S. Steve Corbató corbato@internet2.edu Director, Network Initiatives, Internet2 & Visiting Fellow, Center for
More informationTitle: Collaborative research: End-to-End Provisioned Optical Network Testbed for Large-Scale escience Applications
Year 2Activities report for the NSF project EIN-0335190 Title: Collaborative research: End-to-End Provisioned Optical Network Testbed for Large-Scale escience Applications Date: July 15, 2005 PI: Malathi
More informationStorage Supporting DOE Science
Storage Supporting DOE Science Jason Hick jhick@lbl.gov NERSC LBNL http://www.nersc.gov/nusers/systems/hpss/ http://www.nersc.gov/nusers/systems/ngf/ May 12, 2011 The Production Facility for DOE Office
More informationLawrence Berkeley National Laboratory Lawrence Berkeley National Laboratory
Lawrence Berkeley National Laboratory Lawrence Berkeley National Laboratory Title Scaling the Earth System Grid to 100Gbps Networks Permalink https://escholarship.org/uc/item/80n7w3tw Author Balman, Mehmet
More informationHigh bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK
High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK [r.tasker@dl.ac.uk] DataTAG is a project sponsored by the European Commission - EU Grant IST-2001-32459
More informationDemartek September Intel 10GbE Adapter Performance Evaluation for FCoE and iscsi. Introduction. Evaluation Environment. Evaluation Summary
Intel 10GbE Adapter Performance Evaluation for FCoE and iscsi Evaluation report prepared under contract with Intel Corporation Introduction As the interest in converged networks grows, and as the vendors
More informationAutomated Characterization of Parallel Application Communication Patterns
Automated Characterization of Parallel Application Communication Patterns Philip C. Roth Jeremy S. Meredith Jeffrey S. Vetter Oak Ridge National Laboratory 17 June 2015 ORNL is managed by UT-Battelle for
More informationAdapter Cards ONTAP Fabric-Attached Storage - Current models (page 1 of 5)
Adapter Cards - 9.1 ONTAP Fabric-Attached Storage - Current models (page 1 of 5) Part Number Optical Supported Speeds X113A-R6 [1] 4p 8Gb FC Op - FC: /4/8 Gbps 8.1.1, 8.RC1, 8.3, 9.0RC1, 60 650 800 9000
More informationAdvanced iscsi Management April, 2008
April, 2008 Gene Nagle, istor Networks SNIA Legal Notice The material contained in this tutorial is copyrighted by the SNIA. Member companies and individuals may use this material in presentations and
More informationLow Latency Evaluation of Fibre Channel, iscsi and SAS Host Interfaces
Low Latency Evaluation of Fibre Channel, iscsi and SAS Host Interfaces Evaluation report prepared under contract with LSI Corporation Introduction IT professionals see Solid State Disk (SSD) products as
More informationEnabling a SuperFacility with Software Defined Networking
Enabling a SuperFacility with Software Defined Networking Richard Shane Canon, Tina Declerck, Brent Draney, Jason Lee, David Paul, David Skinner NERSC, Lawrence Berkeley National Laboratory Berkeley, USA
More informationGuide to Networking Essentials, 6 th Edition. Chapter 7: Network Hardware in Depth
Guide to Networking Essentials, 6 th Edition Chapter 7: Network Hardware in Depth Objectives Describe the advanced features and operation of network switches Describe routing table properties and discuss
More informationThe NE010 iwarp Adapter
The NE010 iwarp Adapter Gary Montry Senior Scientist +1-512-493-3241 GMontry@NetEffect.com Today s Data Center Users Applications networking adapter LAN Ethernet NAS block storage clustering adapter adapter
More informationHigh Performance Computing at Mississippi State University
High Performance Computing at Mississippi State University Building Facilities HPC Building - Starkville o CCS, GRI, PET, SimCtr o 71,000 square feet o 3,500 square ft. data center 600 KW generator 60
More informationComparing TCP performance of tunneled and non-tunneled traffic using OpenVPN. Berry Hoekstra Damir Musulin OS3 Supervisor: Jan Just Keijser Nikhef
Comparing TCP performance of tunneled and non-tunneled traffic using OpenVPN Berry Hoekstra Damir Musulin OS3 Supervisor: Jan Just Keijser Nikhef Outline Introduction Approach Research Results Conclusion
More informationTitan - Early Experience with the Titan System at Oak Ridge National Laboratory
Office of Science Titan - Early Experience with the Titan System at Oak Ridge National Laboratory Buddy Bland Project Director Oak Ridge Leadership Computing Facility November 13, 2012 ORNL s Titan Hybrid
More informationNetworks & protocols research in Grid5000 DAS3
1 Grid 5000 Networks & protocols research in Grid5000 DAS3 Date Pascale Vicat-Blanc Primet Senior Researcher at INRIA Leader of the RESO team LIP Laboratory UMR CNRS-INRIA-ENS-UCBL Ecole Normale Supérieure
More informationVIRTUALIZING SERVER CONNECTIVITY IN THE CLOUD
VIRTUALIZING SERVER CONNECTIVITY IN THE CLOUD Truls Myklebust Director, Product Management Brocade Communications 2011 Brocade Communciations - All Rights Reserved 13 October 2011 THE ENTERPRISE IS GOING
More informationThe Titan Tools Experience
The Titan Tools Experience Michael J. Brim, Ph.D. Computer Science Research, CSMD/NCCS Petascale Tools Workshop 213 Madison, WI July 15, 213 Overview of Titan Cray XK7 18,688+ compute nodes 16-core AMD
More informationMidoNet Scalability Report
MidoNet Scalability Report MidoNet Scalability Report: Virtual Performance Equivalent to Bare Metal 1 MidoNet Scalability Report MidoNet: For virtual performance equivalent to bare metal Abstract: This
More informationPerformance Measurement and Evaluation Tool for Large-scale Systems
Performance Measurement and Evaluation Tool for Large-scale Systems Hong Ong ORNL hongong@ornl.gov December 7 th, 2005 Acknowledgements This work is sponsored in parts by: The High performance Computing
More informationGuide to Networking Essentials, 6 th Edition. Chapter 7: Network Hardware in Depth
Guide to Networking Essentials, 6 th Edition Chapter 7: Network Hardware in Depth Objectives Describe the advanced features and operation of network switches Describe routing table properties and discuss
More informationIntroduction to Grid Computing
Milestone 2 Include the names of the papers You only have a page be selective about what you include Be specific; summarize the authors contributions, not just what the paper is about. You might be able
More informationCERN openlab Summer 2006: Networking Overview
CERN openlab Summer 2006: Networking Overview Martin Swany, Ph.D. Assistant Professor, Computer and Information Sciences, U. Delaware, USA Visiting Helsinki Institute of Physics (HIP) at CERN swany@cis.udel.edu,
More information10GE network tests with UDP. Janusz Szuba European XFEL
10GE network tests with UDP Janusz Szuba European XFEL Outline 2 Overview of initial DAQ architecture Slice test hardware specification Initial networking test results DAQ software UDP tests Summary 10GE
More informationHigh-speed networking for scientific applications. John Dennis, NCAR, Robert D. Rusell, UNH, David Starobinski, BU
High-speed networking for scientific applications PhD students: Zhenzhen Yan, Jie Li, Zhengyang Liu, Tian Jin, Zhe Song Collaborators: Chris Tracy, ESnet, Steve Emmerson, UCAR, John Dennis, NCAR, Robert
More informationNetwork and Host Design to Facilitate High Performance Data Transfer
Network and Host Design to Facilitate High Performance Data Transfer Jason Zurawski - ESnet Engineering & Outreach engage@es.net globusworld 2014 April 15 th 2014 With contributions from S. Balasubramanian,
More informationExperiences Harnessing Cray XD1 FPGAs & comparison with other FPGA HPC Systems
Experiences Harnessing Cray XD1 FPGAs & comparison with other FPGA HPC Systems Olaf O. Storaasli, Melissa S. Smith & Sadaf Alam Oak Ridge National Laboratory, Tennessee Status Report Are FPGAs ready for
More information