Joint Techs Masthead
Program for Summer 07 ESCC/Internet2 Joint Techs Workshop

Summer 07 ESCC/Internet2 Joint Techs Workshop

Sunday, July 15, 2007 to Thursday, July 19, 2007
All Times GMT -6

Display Options
Day:
Topic Areas:
Session Details: On Off
(or click on the session title)

Export this page to PDF Printer-friendly version of this doc ical

Sessions currently underway

NOTE: Additional task force and working meetings as well as BoFs and SIGs will be posted as soon as they are confirmed. To view updates please return to this page often.
Sunday, 7/15   Location
11:00 AM - 5:00 PM Registration Desk Open

Atrium 
  Laptop Bar Open

Help Desk Services Area 
1:00 PM - 2:30 PM Click here for live netcast, and video on demand. Campus Networking BoF

   David Farmer , University of Minnesota

Discuss campus networking issues in general. How do we as a community raise the bar on campus networks, since they seem to be a significant part of the end-to-end performance issues?

Auditorium 
2:30 PM - 3:00 PM Break

One West Alcove 
3:00 PM - 4:45 PM Click here for live netcast, and video on demand. Performance Tools Tutorial (includes remote presentations)

   Rich Carlson , Department of Energy, Moderator
   Chin Guok , Energy Sciences Network (ESnet)
   Dimitrios Katramatos , Brookhaven National Laboratory  [htm]  [ppt]
   Andrew Lake , Internet2  [htm]  [ppt]

Rich Carlson will host the 13th Performance Tools Tutorial session. The presenters will provide an update on the development and use of problem-solving tools in campus environments.

Auditorium 
6:00 PM - 8:00 PM Gigapop Geeks BoF

   Jon-Paul Herron , Indiana University
   Dan Magorian , Mid-Atlantic Crossroads (MAX)  [htm]  [ppt]

Hilton Lisle/Naperville
Pine/Spruce 
Monday, 7/16   Location
7:30 AM - 8:30 AM Breakfast

Tent 
7:30 AM - 5:00 PM Registration Desk Open

Atrium 
  Laptop Bar Open

Help Desk Services Area 
8:30 AM - 8:45 AM Click here for live netcast, and video on demand. Welcome / Opening Remarks

   W. Scott Bradley , Brookhaven National Laboratory
   Russ Hobby , Internet2  [htm]  [ppt]
   Don Petravick , Fermi National Accelerator Lab  [pdf]

Auditorium 
8:45 AM - 9:00 AM Click here for live netcast, and video on demand. Local Host Welcome

   Hugh Montgomery , Fermi National Accelerator Lab
   Don Petravick , Fermi National Accelerator Lab

Auditorium 
9:00 AM - 9:15 AM Click here for live netcast, and video on demand. ESnet: Networks for the Future of DOE Science

   William Johnston , Energy Sciences Network (ESnet)  [htm]  [ppt]

DOE's Office of Science is a major funder of physical science research in the US, and the primary funder in several areas. As the scientific experiments get larger and more expensive, the size and scope of the associated collaboration communities increases. Supporting the scientific data analysis of the experiments requires widely distributed data management and processing of massive amounts of data. This, in turn, requires new capabilities and services from the networks that transport the data. This talk will describe the new ESnet architecture and the progress toward a service-oriented WAN environment that is designed to support the new science paradigm.

Auditorium 
9:15 AM - 9:30 AM Click here for live netcast, and video on demand. Internet2 Technical Direction

   Eric Boyd , Internet2  [htm]  [ppt]
   Rick Summerhill , Internet2  [htm]  [ppt]

The Internet2 staff and community are engaged in a wide range of technical efforts. This talk will give an overview of ongoing efforts and the process by which the community can help set the strategic vision for such efforts going forward, with a particular focus on the networking component of cyberinfrastructure.

Auditorium 
9:30 AM - 9:45 AM Click here for live netcast, and video on demand. ESnet Update

   Joseph Burrescia , Energy Sciences Network (ESnet)  [htm]  [ppt]

Update on changes at ESnet in the past six months, especially the ESnet4 transition.

Auditorium 
9:45 AM - 10:00 AM Click here for live netcast, and video on demand. NLR Update

   David Reese , CENIC  [htm]  [ppt]

This session will provide an update on the current NLR services including PacketNet, FrameNet, and WaveNet, as well as the NLRview measurement infrastructure. Examples will be given of how NLR's services, particularly WaveNet and FrameNet, are used by a range of users and research projects to build hybrid networks.

Auditorium 
10:00 AM - 10:15 AM Click here for live netcast, and video on demand. Internet2 Network Update

   Steve Cotter , Internet2  [htm]  [ppt]

This session will provide an update on the recently completed Internet2 Network and its advanced capabilities, as well as progress on the ongoing transition of connectors off the Abilene Network onto the new infrastructure.

Auditorium 
10:15 AM - 10:45 AM Break

Tent 
10:45 AM - 11:00 AM Click here for live netcast, and video on demand. Update on FiberCo/WaveCo Services

   Linda Roos , Internet2  [htm]  [ppt]

This session will provide an overview of FiberCo's dark fiber and professional services offerings, as well as an overview of WaveCo services.

Auditorium 
11:00 AM - 11:20 AM Click here for live netcast, and video on demand. ARIN and Internet Number Resources

   Richard Jimmerson , American Registry for Internet Numbers (ARIN)  [htm]  [ppt]

ARIN will provide an update on Internet number resource distribution and policy. IPv4 depletion and IPv6 transition will be featured in this discussion.

Auditorium 
11:20 AM - 11:40 AM Click here for live netcast, and video on demand. DREN IPv6 Update

   Ron Broersma , Defense Research and Engineering Network (DREN)  [htm]  [ppt]

Auditorium 
11:40 AM - 12:00 PM Click here for live netcast, and video on demand. Using BGP communities to provide desired return-path "hinting"

   Brent Sweeny , Indiana University  [pdf]

A proposal using BGP communities to enable R&E sites to "hint" to their upstream R&E network providers the path they would prefer be used back to them. This could allow those providers to implement generalizable policies to enable those preferences, if desired.

Auditorium 
12:00 PM - 2:00 PM Lunch - Sponsored by Fujitsu

Tent 
12:30 PM - 1:00 PM DWDM GMPLS Deployment Experience (Vendor Soup)

   George Frank , Infinera Corporation  [htm]  [ppt]

Infinera DTN is a DWDM platform with integrated switching capability that supports GMPLS features such as Lambda & Sub-Lambda LSP setup & release, CSPF path computation with explicit routing control, OSPF automatic topology and traffic engineering discovery and routing, end-to-end GMPLS-based recovery, service level management, and UNI, in general mesh topologies. This presentation describes how these features have been deployed in several real networks, and discusses enhancements planned for the future. Topics that will be covered include scaling to large networks, participating in L1VPNs, fast GMPLS recovery techniques, transport layer multipoint connections, circuit bonding, usability and maintenance considerations, and control-plane interoperability with other NEs. The benefits and challenges associated with applications enabled by GMPLS -- such as dynamic bandwidth on demand & L1VPNs -- will be discussed as well.

One East 
12:30 PM - 1:50 PM Bulk Transport Working Group

   Chester Ruszczyk , MIT Haystack Observatory
   Steve Senger , University of Wisconsin-La Crosse

This is a regular meeting of the Bulk Transport Working Group, a collection of individuals addressing the need for a new transport protocol to allow users to determine the degree of speed/loss allowable for their particular application. The group is currently reviewing materials related to Phoebus and other logistical session layer materials developed at the University of Delaware. More information can be found at http://transport.internet2.edu/.

One North 
  Salsa Disaster Planning and Recovery Working Group Meeting

   Joe St Sauver , Internet2/University of Oregon

The Salsa Disaster Recovery (Salsa-DR) working group has been formed to explore and document recommended practices for disaster planning and recovery in higher education. Both current members and any others who may be interested are welcome to attend at this roundtable session. Come and share your disaster planning and recovery concerns and the solutions which you or your institution may be exploring.

One West 
  Click here for live netcast, and video on demand. HENP SIG - Tier 2/3 Science Data requirements

   Rich Carlson , Department of Energy

This SIG will bring LHC scientists and network staff together to review the LHC data requirements. Preliminary indications are that LHC physicists at Tier 3 sites will have bursty network needs (individuals at each site will use 1.25 - 3 Gbps for 2-4 hours every 12-14 days).

Auditorium 
  GPN Technical BoF

   Dale Finkelson , Internet2

Discuss Kansas City network aggregation issues.

Curia II 
12:30 PM - 2:00 PM Wilson Hall tour

The 15th floor of Wilson Hall contains a number of exhibits describing the Fermilab site, accelerator, and experimental apparatus. There is a short movie describing the nature of the science performed at the lab. Good views of the lab site make this a worthwhile stop. Visitors must be accompanied to the 15th floor, but may return to the atrium level at any time. The tour will start when a sufficient group has gathered. Allow about 15 minutes.

meet at the registration desk 
  Collider Detector Facility (CDF) tour

CDF and its companion experiment, DZero, are the two large experiments/detectors studying the colliding proton and anti- proton beams of the Fermilab Tevatron. This tour will visit the CDF Assembly Hall. Although the detector itself is in position within the Collision Hall and thus not visible, the Assembly Hall surroundings and displays will give a good idea of the complexity of the experiment. CDF is a short bus ride around the accelerator ring from Wilson Hall. Tour groups will be limited to about 10 people, and two groups can fit into the one bus which will make the trip to CDF. Allow about 45 minutes.

meet in front of Wilson Hall 
  Grid Computing Center (GCC) tour

As the extent of its computing facilities has grown, the Fermilab Computing Division has expanded from its Feynman Computing Center (FCC) home into a "recycled and expanded" former experimental area, the GCC. This facility now supports the bulk of Fermilab's computing capacity, with over 3000 compute nodes and over 10 petabytes of robotic mass storage capacity, all interconnected with high-speed networks to the remainder of the Fermilab site and the external world. GCC is a short bus ride from Wilson Hall. Each of two simultaneous tour groups sharing the bus will be limited to about 10 people. Allow about 45 minutes.

meet in front of Wilson Hall 
1:15 PM - 1:45 PM Delivering Intelligent Ethernet Services to the Research & Education Community (Vendor Soup)

   Fred Ellefson , ADVA Optical Networking

As Ethernet services scale, the requirement for demarcation devices between end users and research and education national backbone networks is key to providing intelligent Ethernet services that support remote testing, monitoring, and service level agreements (SLAs). ADVA will discuss new technology, standards, and operations methodology that many service providers are using to roll out their carrier Ethernet services, with an emphasis on Ethernet operations, administration and maintenance (OAM) capabilities, their use in demarcation devices, and how they can migrate to the research and education environment.

One East 
2:00 PM - 2:20 PM Click here for live netcast, and video on demand. Measurement Overview

   Eric Boyd , Internet2  [htm]  [ppt]
   Joe Metzger , Energy Sciences Network (ESnet)  [htm]  [ppt]

The Internet2 and ESnet community has a suite of performance tools available for monitoring networks. We are well on the way to having a suite of performance middleware for exchanging and discovering data across multiple administrative boundaries. This talk will motivate the need to deploy such tools and infrastructure across the R&E community.

Auditorium 
2:20 PM - 2:40 PM Click here for live netcast, and video on demand. Continuous Throughput Testing and Dynamic Visualization Utilizing Network Diagnostic Tool (NDT)

   Carla Hunt , MCNC  [htm]  [ppt]
   Andrew Mabe , MCNC  [htm]  [ppt]

MCNC had a need to accurately test the bandwidth for hundreds of hosts across various networks back into the North Carolina Research and Education Network (NCREN). Due to the constraint that MCNC could not install software at the remote locations, we chose to expand on the NDT tool. What we have developed automatically repeats NDT bandwidth tests, analyzes results, and graphs and displays the information for each host. This solution is built by using the modified NDT tool, RRD archiving and PHP.

Auditorium 
2:40 PM - 3:00 PM Click here for live netcast, and video on demand. NetFlow Tools and Analysis at Fermilab

   Andrey Bobyshev , Fermi National Accelerator Lab  [htm]  [ppt]
   Phil DeMar , Fermi National Accelerator Lab  [htm]  [ppt]

In this presentation we are going to talk about various tools based on analysis of NetFlow data that are used at Fermilab for traffic characterization and network performance estimation at USCMS Tier 1 center, other experiments, and entire site network. We are using NetFlow data gathered from multiple local and border routers for checking of traffic consistency, End-To-End circuit and Policy Based Routing, nearly real-time detection of onsite and offsite scanning, and automated blocking of most aggressive scanners. Tools and results of analysis are going to be presented and discussed.

Auditorium 
3:00 PM - 3:20 PM Click here for live netcast, and video on demand. Wide Area Network Performance Analysis Methodology

   Mark Bowden , Fermi National Accelerator Lab  [htm]  [ppt]
   Phil DeMar , Fermi National Accelerator Lab  [htm]  [ppt]
   Wenji Wu , Fermi National Accelerator Lab  [htm]  [ppt]

The computing models for LHC experiments are globally distributed and grid-based. In such a computing model, the experiments' data must be reliably and efficiently transferred from CERN to Tier 1 regional centers, processed, and distributed to other centers around the world. Obstacles to good network performance arise from many causes and can be a major impediment to the success of this complex, multi-tiered data grid. Factors that affect overall network/application performance exist on the network end systems themselves (application software, operating system, hardware), in the local area networks that support the end systems, and within the wide area networks. Since the computer and network systems are globally distributed, it can be very difficult to locate and identify the factors that are hurting application performance. In this paper, we present an end-to-end network/application performance troubleshooting methodology developed and in use at Fermilab. The core of our approach is to narrow down the problem scope with a divide and conquer strategy. The overall complex problem is split into two distinct sub-problems: network end system diagnosis and tuning, and network path analysis. After satisfactorily evaluating, and if necessary resolving, each sub-problem, we conduct end-to-end performance analysis and diagnosis. The paper will discuss tools we use as part of the methodology. The long term objective of the effort is to enable end users to conduct much of the troubleshooting themselves, before (or instead of) calling upon network and end system "wizards", who are always in short supply.

Auditorium 
3:20 PM - 3:40 PM Click here for live netcast, and video on demand. Performance Update

   Jeff Boote , Internet2  [htm]  [ppt]
   Eric Boyd , Internet2

Internet2 is involved in several exciting projects related to performance and measurement. Eric Boyd will provide an update on the progress of:
- perfSONAR, a network monitoring infrastructure that makes it easier to solve end-to-end performance problems on paths crossing several networks; contributors to perfSONAR development include Internet2, G√ČANT2, and ESnet members, among others. Demonstrations of perfSONAR included in the presentation illustrate the "service-oriented" aspects of perfSONAR by showing how disjoint services such as Measurement Points (MPs), Measurement Archives (MAs), and the Lookup Service (LS) can interact via two graphical displays (an interactive map of utilization data and a configurable display of network traffic).
- Performance tool development, including updates on OWAMP, Bandwidth Test Controller, Network Diagnostic Tool, and Thrulay, as well as the development of a Knoppix disk with downloads of these and other tools in an easy-to-use format.
- Network Performance Workshops, which have been offered in 16 locations and are scheduled for several more in upcoming months.
- Application community-specific performance efforts, including a series of workshops with the HENP Large Hadron Collider community, the performing arts community, and state networks related to meeting measurement and performance needs. More information on all of these topics can be found at http://e2epi.internet2.edu/.

Auditorium 
3:40 PM - 4:10 PM Break

Tent 
4:10 PM - 4:30 PM Click here for live netcast, and video on demand. perfSONAR demonstration

   Jeff Boote , Internet2  [htm]  [ppt]
   Jason Zurawski , Internet2  [htm]  [ppt]

This session will show the utility of exporting network monitoring data using the perfSONAR framework. The demo will access performance statistics across multiple administrative domains including universities, national labs and backbone networks.

Auditorium 
4:30 PM - 4:50 PM Click here for live netcast, and video on demand. Using perfSONAR Tools in a Production Environment

   Marian Garcia Vidondo , DANTE  [htm]  [ppt]

This talk will introduce how the End to End Coordination Unit (E2ECU) is using the End to End Monitoring System (E2Emon) to monitor the Large Hadron Collider Optical Private Network (LHCOPN).

Auditorium 
4:50 PM - 5:10 PM Click here for live netcast, and video on demand. Heterogeneous Data Plane Testing on USN, ESnet, and Internet2 Networks

   Nasir Ghani , Tennessee Tech University  [htm]  [ppt]
   Chin Guok , Energy Sciences Network (ESnet)  [htm]  [ppt]
   Tom Lehman , University of Southern California  [htm]  [ppt]
   John Moore , ITEC - North Carolina  [pdf]
   Nagesware Rao , Oak Ridge National Laboratory  [htm]  [ppt]
   John Vollbrecht , Internet2  [htm]  [ppt]

Multi-layer, multi-domain hybrid network architectures present several new issues due to the heterogeneous nature of the data plane. These networks provide the ability to establish paths (circuits) at various technology levels such as lambda, SONET, Ethernet, and router MPLS. This talk presents an overview of an effort to evaluate and characterize data paths constructed at the various technology layers as well as data paths constructed via the concatenation of multiple technology-diverse segments. The approach is to combine data collection with advanced simulation and modeling capabilities to predict the performance of existing hybrid networks and also assist in the design of alternative hybrid network architectures. This work is utilizing the infrastructures of UltraScience Net, ESnet, and Internet2 for data collection and as the basis for the analysis work.

Auditorium 
6:00 PM - 8:00 PM Reception - Sponsored by Ciena, Juniper, and Internet2

Hilton Lisle/Naperville
Maple/Elm/Ash 
7:30 PM - 7:50 PM PGP Key Signing

   Doug Pearson , Indiana University

Hilton Lisle/Naperville
Atrium across from Maple 
8:30 PM - 10:00 PM Pilot BOF

The Pilot BoF was originally started by Alan Whinery at the Vancouver JTW, as a means of paying his debt to society. It has done practically none of that so far. If you are an pilot, current or not, or interested in being an pilot, or not, you are welcome. (Welcomeness is very broad.) If your attitude towards piloting is ambivalent, you will be counted as "interested". In this case, a "pilot" refers to someone who possesses a certification from a national government which certifies that they may legally fly an airplane, balloon, sailplane, airship, or rotorcraft. We would also accept harbor or river pilots, etc. (that would be COOL)

Hilton Lounge 
Tuesday, 7/17   Location
7:30 AM - 8:30 AM Breakfast

Tent 
7:30 AM - 5:00 PM Registration Desk Open

Atrium 
  Laptop Bar Open

Help Desk Services Area 
8:30 AM - 8:50 AM Click here for live netcast, and video on demand. Trends in IPv6

   Bill Cerveny , Arbor Networks  [htm]  [ppt]

The nature of IPv6 has evolved since it was first introduced around 1996. The intention of this presentation is to review some of the changes that have evolved in IPv6 from the perspective of how it can or should be implemented. It is hoped that this presentation can serve as something of an update for someone who learned about IPv6 anywhere from two to six years ago, but has not been actively following the technology.

Auditorium 
8:50 AM - 9:10 AM Click here for live netcast, and video on demand. IPv6 Support in Windows Vista

   Sean Siler , Microsoft  [pdf]

Windows Vista provides more and better IPv6 integration than its predecessors. This talk will provide an overview of the IPv6 functionality of Vista and its bundled applications.

Auditorium 
9:10 AM - 9:40 AM Click here for live netcast, and video on demand. Hybrid Network Control Plane Interoperation Between Internet2 and ESnet

   Chin Guok , Energy Sciences Network (ESnet)  [htm]  [ppt]
   Andrew Lake , Internet2  [htm]  [ppt]
   Tom Lehman , University of Southern California  [htm]  [ppt]
   John Vollbrecht , Internet2  [htm]  [ppt]

An important capability for hybrid networks will be the ability to dynamically set up paths on a multi-domain, multi-technology basis. This presentation will provide an update on a joint effort between Internet2 and ESnet to develop an Inter-Domain Controller which will enable this type of multi-domain provisioning. This includes combining technologies developed as part of the OSCARS and DRAGON projects to allow Internet2 HOPI/DCS and ESnet IP/SDN networks to interoperate. Status of our efforts to complete similar interoperability testing with GEANT will also be presented.

Auditorium 
9:40 AM - 10:05 AM Click here for live netcast, and video on demand. CHEETAH Applications and Control-Plane Testing on HOPI (includes remote presentation)

   Tao Li , University of Virginia
   Malathi Veeraraghavan , University of Virginia  [htm]  [ppt]

This talk will describe our experiments with immediate-request (IR) applications, such as web proxy/caching services, for dynamic circuit services, and CHEETAH control-plane software on the HOPI testbed.

Auditorium 
10:05 AM - 10:35 AM Break

Tent 
10:35 AM - 11:00 AM Click here for live netcast, and video on demand. TeraPaths: Establishing End-to-End QoS Paths through L2 and L3 WAN Connections (remote presentation)

   Dimitrios Katramatos , Brookhaven National Laboratory  [htm]  [ppt]

The TeraPaths project at BNL investigates the combination of DiffServ-based LAN QoS with WAN MPLS tunnels and dynamic circuits in establishing end-to-end (host-to-host) virtual paths with guaranteed bandwidth. These virtual paths prioritize, protect, and throttle network flows in accordance with site agreements and user requests, and prevent the disruptive effects that conventional network flows can cause in one another.

Auditorium 
11:00 AM - 11:20 AM Click here for live netcast, and video on demand. Infiniband in the Data Center

   Steven Carter , Oak Ridge National Laboratory  [htm]  [ppt]
   Makia Minich , Oak Ridge National Laboratory  [htm]  [ppt]
   Nagesware Rao , Oak Ridge National Laboratory  [htm]  [ppt]

High-performance file systems that span across supercomputers and storage systems are becoming increasingly important, as these resources are being dispersed both within the enterprise and across the wide-area. Infiniband has been proposed as a candidate to fill the gap left by other technologies in filling this need. In this presentation, we present an experimental study of Infiniband transport over ORNL infrastructure of Cray XT3 supercomputer and cluster storage system for local-area and DOE UltraScience Network (USN) for wide-area. We also discuss the positive and negative aspects of deploying Infiniband as a data center interconnect for the local- and wide-area.

Auditorium 
11:20 AM - 11:40 AM Click here for live netcast, and video on demand. Packet-Centric Transport

   Enrique Hernandez-Valencia , Lucent Technologies  [htm]  [ppt]

As optical transport networks evolve to a hybrid WDM/TDM/packet transport model able to support packet-centric operations, several key issues will need to be addressed. Chief among these issues are the class of switching fabric that is used in the transport device, and the forwarding framework for packet operations. Should the fabric be a hybrid circuit/packet switching fabric or a full packet switching fabric? Should packet forwarding be based on an L2 or an L3 framework? This session will discuss the tradeoffs of these options and present what a leading candidate solution might be. Control plane issues and solutions for evolving the current generation of optical networks to a packet-centric optical network will also be discussed.

Auditorium 
11:40 AM - 12:00 PM Click here for live netcast, and video on demand. 40 Gbps and 100 Gbps WAN Transport Technologies for Research & Education Networks

   Tom McDermott , Fujitsu Network Communications  [htm]  [ppt]

Research and Education (R&E) networks provide the infrastructure that allows researchers to experiment with new state-of-the-art technologies such as High Performance Computing (HPC), remote visualization, and support of the Large Hadron Collider (LHC). Each of these technologies depends on very high capacity optical transmission facilities, and the potential of 40 Gbps and eventually 100 Gbps transmission speeds offer the capacity to enable new types of research experiments. Optical transmission systems will be enhanced to provide these services to R&E networks, often using existing provider networks and fibers. This presentation discusses the optical transmission technologies needed to enhance current optical systems to provide these data rates, and the impact that higher transmission rates will have on existing fiber systems and networks.

Auditorium 
12:00 PM - 2:00 PM Lunch - Sponsored by Qwest

Tent 
12:30 PM - 1:00 PM Reliable Networking - New Innovations (Vendor Soup)

   Force 10 Representative   , Force 10 Networks

With new VoIP, network based services, security and mobility requirements, reliable networking is more important than ever. And that reliability needs to extend all the way from the wide area network to the LAN and the wiring closet. In this session we will discuss newly announced and available Force10 Networks switch products for Gigabit Ethernet, 10 GbE, and power over ethernet (PoE), plus low latency 10 GbE implementations and the technologies and innovations behind these new products.

One East 
12:30 PM - 1:50 PM perfSONAR Working Group

   Jeff Boote , Internet2
   Martin Swany , University of Delaware

This is a regular meeting of the perfSONAR Working Group, a collection of individuals deploying the perfSONAR measurement infrastructure and/or developing tools for use with the international perfSONAR framework. The group meets regularly by phone; this is an opportunity for it to meet face-to-face as well as include new potential working group members. More information can be found at http://www.perfsonar.net/.

Curia II (2nd Floor) 
  Multicast "Ask the Experts"

   Alan Crosswell , Columbia University

One West 
  MREN/CIC Chicago BoF

   Patrick Christian , University of Wisconsin
   Linda Winkler , Argonne National Laboratory

Meeting of MREN and CIC techs, discussing Chicago network aggregation issues.

RaceTrack (7th Floor) 
  Click here for live netcast, and video on demand. IPv6 Support in Windows Vista BoF

   Sean Siler , Microsoft  [pdf]

Auditorium 
12:30 PM - 2:00 PM Wilson Hall tour

The 15th floor of Wilson Hall contains a number of exhibits describing the Fermilab site, accelerator, and experimental apparatus. There is a short movie describing the nature of the science performed at the lab. Good views of the lab site make this a worthwhile stop. Visitors must be accompanied to the 15th floor, but may return to the atrium level at any time. The tour will start when a sufficient group has gathered. Allow about 15 minutes.

meet at the registration desk 
  Collider Detector Facility (CDF) tour

CDF and its companion experiment, DZero, are the two large experiments/detectors studying the colliding proton and anti- proton beams of the Fermilab Tevatron. This tour will visit the CDF Assembly Hall. Although the detector itself is in position within the Collision Hall and thus not visible, the Assembly Hall surroundings and displays will give a good idea of the complexity of the experiment. CDF is a short bus ride around the accelerator ring from Wilson Hall. Tour groups will be limited to about 10 people, and two groups can fit into the one bus which will make the trip to CDF. Allow about 45 minutes.

meet in front of Wilson Hall 
  Grid Computing Center (GCC) tour

As the extent of its computing facilities has grown, the Fermilab Computing Division has expanded from its Feynman Computing Center (FCC) home into a "recycled and expanded" former experimental area, the GCC. This facility now supports the bulk of Fermilab's computing capacity, with over 3000 compute nodes and over 10 petabytes of robotic mass storage capacity, all interconnected with high-speed networks to the remainder of the Fermilab site and the external world. GCC is a short bus ride from Wilson Hall. Each of two simultaneous tour groups sharing the bus will be limited to about 10 people. Allow about 45 minutes.

meet in front of Wilson Hall 
1:15 PM - 1:45 PM 1696 ROADM (Vendor Soup)

   Ben Bacque , Alcatel-Lucent

The Alcatel-Lucent 1696 ROADM platform offers significant scalability to fiber networks while solving several key issues associated with such high-capacity networks. One of the key benefits the 1696 ROADM product offers is the ability to monitor and manage wavelengths without needing intrusive procedures. This is accomplished using our patented Wave Tracker technology, which allows an operator to point and click "any to any" wavelength reassignment, run expansion and failure scenarios before they happen, and monitor the network down to the wavelength and module level. The second issue of reconfigurability and scalability can be solved from an architectural perspective using Reconfigurable Optical Switching, which also provides the flexibility to start with 10 Gbps wavelengths and migrate to 40 Gbps and above without losing any operational benefits. This session will cover the architecture, features and benefits of the Alcatel-Lucent 1696 ROADM product.

One East 
2:00 PM - 2:20 PM Click here for live netcast, and video on demand. The Lambda Station Project

   Andrey Bobyshev , Fermi National Accelerator Lab  [htm]  [ppt]
   Phil DeMar , Fermi National Accelerator Lab  [htm]  [ppt]
   Maxim Grigoriev , Fermi National Accelerator Lab  [htm]  [ppt]

Lambda Station is a research project of Fermilab and Caltech. The goal of this project is to design, develop, and investigate network services for path selection, admission control, and flow-based forwarding of traffic for data-intensive Grid applications. Lambda Station deals with the last-mile problem in local area networks, connecting production clusters through a rich array of wide area networks. Selective forwarding of traffic is controlled dynamically at the demand of applications. In the talk we will introduce design principles, details of Java-based implementation, and integration of Lambda Station client code with the SRM/dCache storage system. Also we will present the results of using the Lambda Station service with production clusters at Fermilab to transfer data to Caltech over UltraLight, an advanced R&D network.

Auditorium 
2:20 PM - 2:40 PM Click here for live netcast, and video on demand. Dynamic Lightpath Connectivity Solutions for Research Networks

   Jeff Verrant , Ciena  [htm]  [ppt]

Ciena will provide an overview of technology drivers and building blocks for dynamic optical network transport and switching. Additionally, key capabilities such as OTN, multi-protocol support, and automated control plane technologies will be explained in context of providing automated provisioning of lightpaths to meet regional and global research networking requirements in hybrid network architectures. An update will be provided on current implementations underway globally.

Auditorium 
2:40 PM - 3:00 PM Click here for live netcast, and video on demand. Reconfigurable DWDM

   Nazar Neayem , Alcatel-Lucent  [htm]  [ppt]

While DWDM technology has provided significant scalability to fiber networks, several key issues still remain. Fiber networks are still difficult to manage and troubleshoot during fault conditions without performing intrusive testing procedures. Also, they are relatively static in their configuration and cannot be dynamically reconfigured or scaled. The first issue of wavelength management and monitoring can be solved using Wave Tracker technology. This technology can allow an operator to point and click "any to any" wavelength reassignment, run expansion and failure scenarios before they happen, and monitor the network down to the wavelength and module level. The second issue can be solved from an architectural viewpoint using ROADM, which provides the flexibility to, say, start with 10Gbps wavelengths and migrate to 40Gbps and above without losing any operational benefits. This session will cover these technologies, which are going to be prevalent in next-generation optical networks.

Auditorium 
3:00 PM - 3:20 PM Click here for live netcast, and video on demand. Anatomy of Ethernet Switch/Router Resiliency

   Force 10 Representative   , Force 10 Networks  [htm]  [ppt]

As Ethernet switch/routers continue to scale in terms of link speed and port density, device resiliency is becoming an indispensable system attribute. The best approach to maximizing the reliability/resiliency and stability of switch/routers involves the application of three basic principles:
- Elimination of single points of failure
- Ensuring that any failures that do occur are constrained to affect only one system component or subsystem
- Ensuring that when a subsystem fails, becomes compromised, or needs updating, the recovery can be accomplished without disrupting the continued operation of the overall system.
These principles are the basis for a resiliency architecture, which allows switch/routers to continue network operations in the face of the full gamut of fault conditions, including the failure of hardware components, software faults and restarts, link failures, protocol restarts, and attempts by intruders to disrupt normal traffic flow. This presentation will discuss high availability features and protocols for resiliency of switch/routers in your networks.

Auditorium 
3:20 PM - 3:40 PM Click here for live netcast, and video on demand. Complexity and Degrees of Freedom in Network Design

   Michael Sinatra , University of California, Berkeley  [htm]  [ppt]

This presentation is a longer prequel to my Lightning Talk at the Winter 2007 JT in Minneapolis ("Why we should eighty-six five nines"). Network designs are becoming increasingly complex, not only because more tangential "services" (firewalling / packet filtering, packet shaping, content acceleration, intrusion prevention, and the like) are being incorporated into network designs, but also, ironically, because network designers are trying to engineer higher availability into networks. I draw on academic literature on complex systems theory and normal accidents to remind network engineers that placing more devices and services into networks, even if such services are "redundant" or "highly available", necessarily adds complexity to network designs. Ironically, attempts to engineer greater reliability into networks can often have the opposite effect. Moreover, they can create hard-to-diagnose problems that wreak havoc with user expectations and require expensive and time-consuming troubleshooting. I draw a number of conclusions, and I stress the need for greater elegance and simplicity in designs. Moreover, I suggest that an attractive option may be to forgo "five-nines"-style redundancy in exchange for designs that are simpler and easier to maintain.

Auditorium 
3:40 PM - 4:10 PM Break

Tent 
4:10 PM - 4:30 PM Click here for live netcast, and video on demand. Experiences with Deploying a 10G+ Network

   Steven Carter , Oak Ridge National Laboratory  [pdf]
   Josh Lothian , Oak Ridge National Laboratory  [pdf]

Many infrastructure upgrades are required when preparing to deploy a peta-scale supercomputer. Among these upgrades is the network. In order to satisfy the data transfer requirements of this size supercomputer, large amounts of local- and wide-area bandwidth are needed. Furthermore, these upgrades need to be done in a cost-effective manner with little or no impact on current operations, while providing sufficient capacity and robustness to satisfy the Center's needs. This presentation details the Oak Ridge National Laboratory Center for Computational Science's road from a 1G to a 10G+ network. It will explore the process from design to testing and deployment, and highlight lessons learned.

Auditorium 
4:30 PM - 4:50 PM Click here for live netcast, and video on demand. "Client-side DWDM": A Model for Next-Generation Baltimore Region Optical Fanout

   Dan Magorian , Mid-Atlantic Crossroads (MAX)  [htm]  [ppt]

Seven participants in the Baltimore region optical consortium are collaborating to build BERnet's next-generation optical fanout this summer. The scheme uses custom 40-channel filter pairs over existing metro dark fiber segments, lit by DWDM SFPs and XFPs in the client ports of the backbone DWDM system, and measured by external optical power monitors from the filter tap ports. This "DIY optical" scheme is extremely economical relative to full integrated DWDM platforms, yet keeps multiple transponder pair costs down and does not have the disadvantages of "alien waves" into the core. It is easily extensible to many other fiber situations as well.

Auditorium 
6:00 PM - 8:00 PM RONs/Connectors BoF

   Ana Hunsinger , Internet2
   Heather Todorov , Internet2

Hilton Lisle/Naperville
Maple/Elm/Ash 
8:15 PM - 10:15 PM JET Meeting (invitation only)

   Paul Love , NCO and NOAA

The Joint Engineering Team is concerned with the engineering coordination of the various US Federal R&E networks, joined by invited academic networks. While this is a by-invitation meeting, the JET is receptive to requests to attend. Please send a note to Paul Love at love@nitrd.gov if you'd like to attend. As a reference point, the JET's typical event horizon is fairly short (6-12 months) but on occasion it looks out to the mid-term.

Hilton Lisle/Naperville
Birch/Hickory 
Wednesday, 7/18   Location
7:30 AM - 8:30 AM Breakfast

Tent 
7:30 AM - 4:30 PM Registration Desk Open

Atrium 
  Laptop Bar Open

Help Desk Services Area 
8:30 AM - 8:50 AM Click here for live netcast, and video on demand. CMS: Priming the Network for LHC Startup and Beyond

   Oliver Gutsche , Fermi National Accelerator Lab  [pdf]

The startup of LHC data-taking is approaching quickly. To enable its over 2000 collaborators to analyze data taken by the detector, CMS is currently preparing and exercising all its systems to be ready for day one. Due to the distributed nature of the computing model and the need to split the data samples geographically and provide access to them, CMS relies on a very powerful, fast and reliable network infrastructure to distribute its data. This talk will discuss the computing and analysis model of CMS, and will describe the important role of the network at the various levels of the CMS tiered computing infrastructure. It will present the ongoing efforts for preparation and commissioning and will also describe the standard running scenario during stable data-taking of CMS/LHC.

Auditorium 
8:50 AM - 9:10 AM Click here for live netcast, and video on demand. Federated Model for Cyber Security

   Scott Pinkerton , Argonne National Laboratory  [pdf]

We are promoting and developing an infrastructure that allows organizations to share information related to cyber-security activities, with a goal of improving our operational response to incidents. This starts with establishing trust relationships with at least one external organization, and then sharing data. Typical data being shared would be info about external IP addresses, relative to their site, that have been hostile. The who, what, where, and when -- plus any information about previous malicious activity from this same IP -- is distributed within the federation. This information is packaged up and distributed using the infrastructure we are creating (a limited-function web site that supports a push and pull of data from members of the federation). An additional capability that the infrastructure supports is the ability to ask a limited number of questions from federation members in a near real-time manner. E.g. have you seen port 80 traffic from this IP; have you seen outbound port 21 traffic to this IP; is this a live/valid IP address at your site today? This infrastructure also supports actions like "strongly suggest blocking outbound port 80 traffic to this IP."

Auditorium 
9:10 AM - 9:30 AM Click here for live netcast, and video on demand. Introduction to the US Government DNSSEC Pilot

   Scott Rose , NIST  [htm]  [ppt]

The Secure Naming Infrastructure Pilot (SNIP) is a new pilot program for DNSSEC deployment. The primary focus is on US government agencies who need to deploy DNSSEC to meet FISMA requirements, but the project is open to anyone. This presentation will be a brief overview of the project and its goals of aiding deployment and testing of DNSSEC enabled software.

Auditorium 
9:30 AM - 9:50 AM Click here for live netcast, and video on demand. Lawful Intercept and Advanced Networking Environments

   Joe St Sauver , Internet2/University of Oregon  [pdf]

CALEA, the United States "Communication Assistance for Law Enforcement Act", requires entities covered by the Act to be ready to respond to law enforcement requests for lawful intercepts should such a request be received. While CALEA has traditionally been targeted at voice providers, recent changes have extended that statute to also include facilities-based broadband Internet providers (including at least some universities and regional networks.) When applying CALEA to higher education's network environment, our architectures, protocols, and advanced capabilities can all differ in significant ways from what's commonly deployed in the commercial Internet. This talk will discuss those differences and some of the practical implications associated with them, offering recommendations for addressing the resulting challenges.

Auditorium 
9:50 AM - 10:10 AM Click here for live netcast, and video on demand. Defense-in-Depth using Network Virtualization and Network Admission Control

   Steven Carter , Oak Ridge National Laboratory  [htm]  [ppt]
   Susan Stewart , Cisco Systems  [htm]  [ppt]

Changing security requirements are making it necessary to refocus network security from being predominately border-oriented to being ingrained in the very fabric of the network. Network Virtualization and Network Admission Control allow the network administrator to create credible, virtual networks that separate both the data and control planes. This security can extend all the way to the desktop to both authenticate its user and validate its security posture. This presentation will give a high-level overview of these techniques and how they can be deployed into the type of exception-filled, heterogeneous networks characteristic of today's research and education institutions.

Auditorium 
10:10 AM - 10:40 AM Break

Tent 
10:40 AM - 11:00 AM Click here for live netcast, and video on demand. Computer Security at Fermilab

   Joe Klemencic , Fermi National Accelerator Lab
   Andy Rader , Fermi National Accelerator Lab  [pdf]
   Tim Rupp , Fermi National Accelerator Lab  [pdf]

Commercial and in-house tools for networking and security will be discussed.

Auditorium 
11:00 AM - 11:20 AM Click here for live netcast, and video on demand. Cyberinfrastructure: What Is It?

   Russ Hobby , Internet2  [htm]  [ppt]

"Cyberinfrastucture" (CI) is the current buzzword for information technology, but what is it? This presentation will provide one view of the various parts and players in the CI picture.

Auditorium 
11:20 AM - 11:50 AM Click here for live netcast, and video on demand. Lightning Talks

   Matt Crawford , Fermi National Accelerator Lab  [htm]  [ppt]
   David Farmer , University of Minnesota  [htm]  [ppt]
   Dale Finkelson , Internet2
   Maxim Grigoriev , Fermi National Accelerator Lab  [htm]  [ppt]
   Alan Whinery , University of Hawaii  [htm]  [ppt]

A Lightning Talk is a very short (no more than 5 minutes) talk on a topic of your choice -- anything interesting or timely, or even a bit off the wall. Submissions will be solicited during the workshop. Please send proposals for Lightning Talks to jt-lightning-talks@internet2.edu.

Slides are optional. All slides must be pre-loaded on the podium laptop -- you will not be permitted to use your own laptop, as there just won't be time for swapping. Slides must be received by 9:30 AM on Wednesday.

Remember, the 300-second limit will be strictly enforced! Some tips on how to do a successful Lightning Talk are at http://www.perl.com/pub/a/2004/07/30/lightningtalk.html.

1."Importance of the unified XML schema and messaging model for dynamic circuit reservation and QOS projects" - Maxim Grigoriev, FNAL

2. "v6 Workshops" - Dale M Finkelson Unv, N-Lincoln

3. "Managing your Network using LinkedIn" - David Farmer, Unv MInn

4. "Phone Message-Based Control And Blogging: An Implementation" - Alan Whinery, Unv. Hawaii

5. "Reducing RTT with novel transmission" - Matt Crawford, FNAL

Auditorium 
11:50 AM - 12:00 PM Click here for live netcast, and video on demand. Closing Remarks

   W. Scott Bradley , Brookhaven National Laboratory
   Russ Hobby , Internet2

Auditorium 
12:00 PM - 2:00 PM Lunch

Tent 
12:30 PM - 1:00 PM Ciena CN4200 and CoreDirector solutions for R&E (Vendor Soup)

   Jeff Verrant , Ciena

Ciena will provide an overview of technology drivers and building blocks for dynamic optical network transport and switching. Additionally, key capabilities such as OTN, multi-protocol support, and automated control plane technologies will be explained, in context of providing automated provisioning of lightpaths to meet regional and global research networking requirements in hybrid network architectures. An update will be provided on current implementations underway globally.

One East 
12:30 PM - 1:50 PM Northern Tier Engineering Group BoF

   David Farmer , University of Minnesota

Discuss the engineering of the Northern Tier Project.

Auditorium 
1:00 PM - 4:30 PM ESCC Meeting (separate registration required)

   W. Scott Bradley , Brookhaven National Laboratory

See http://jointtechs.es.net/Illinois2007/ESCC-July-07-Agenda.doc.

One West 
2:30 PM - 2:45 PM ESCC Break

Tent 
Thursday, 7/19   Location
8:00 AM - 9:00 AM ESCC Breakfast

serving in One West Alcove,
seating in One North 
9:00 AM - 4:00 PM ESCC Meeting (separate registration required)

   W. Scott Bradley , Brookhaven National Laboratory

See http://jointtechs.es.net/Illinois2007/ESCC-July-07-Agenda.doc.

One West 
10:30 AM - 10:45 AM ESCC Break

serving in One West Alcove,
seating in One North 
12:00 PM - 1:00 PM ESCC Lunch

serving in One West Alcove,
seating in One North 
3:00 PM - 3:15 PM ESCC Break

serving in One West Alcove,
seating in One North 

Last Updated: Wednesday, July 18, 2007 1:21 PM