Fall 2010 Internet2 Member Meeting

Use Internet2 SiteID

Already have an Internet2 SiteID?
Sign in here.

Internet2 SiteID

Poster Sessions

On Tuesday and Wednesday of the Fall 2010 Internet2 Member Meeting in the Garden Court foyer on the main meeting floor, presenters and organizational representative interacted and discussed their posters during the following times:

Tuesday November 2 10:00 - 10:30 AM 2:30 - 3:00 PM 4:00 - 4:30 PM
Wednesday November 3 10:00 - 10:30 AM 2:30 - 3:00 PM 4:00 - 4:30 PM
Thursday November 4 10:00 - 10:30 AM


  • ADVA Optical Networking
  • BTI Systems
  • Ciena
  • Ekinops
  • Fujitsu Network Communications
  • Infinera
  • Juniper Networks
  • National Cancer Institute
  • National Institute of Informatics
  • Ohio University
  • University of Tennessee

TERENA Task Force on Media Production, Management and Distribution

Providing recorded or live multimedia footage of university lectures to the education community has become increasingly popular. To attract students, universities often differentiate themselves by providing access to on-line lectures and content outside of the university. The National Research and Education Networks (NRENs) are in a good position to provide multimedia recording, management, repository and distribution services to universities - they understand the special requirements of the academic and research community. TERENAÕs new task force on media management, TF-Media, brings together European NRENs, universities, librarians and other academic content owners to explore the architectures, technologies and workflow processes necessary for various rich media content production management and distribution. The poster gives more information about the details of the activity.

  • Peter Szegedi - TERENA

100G and OTN

As the thirst for bandwidth continues, the drive for increasing raw capacity continues. Recent advances in optical components and electronics has led to the introduction of commercially available 100Gbps transport systems. These systems are now in use in enterprise applications as well as R&E networks. But transport capacity alone is not enough to meet the challenge. Optical Transport Networking (OTN) offers a set of standards that define how capacity at higher rates, such as 40 or 100Gbps and beyond, is utilized by any underlying service protocol. OTN provides a wrapper technique which encapsulates various protocols, such as SONET/SDH, or Ethernet onto a unified format while also providing performance mechanisms such as forward error correction and performance monitoring.

  • John Lankford - Ciena

The Role of Optical Transport in Cyberinfrastructure

This session will discuss the role of optical transport in support of cyberinfrastructure. The poster session will discuss transport technologies that help to reduce the cost of efficiently connecting campuses to regional networks and Internet2.

  • Rob Adams - Ekinops

REFEDs The R&E Identity Federations forum

With the growing number and broadening deployment of identity federations in Europe and beyond, REFEDs has become a truly global forum since its establishment in 2005, with regular participation from five continents. REFEDs addresses mostly policy-related topics about federation e-identity, but also includes technical and outreach topics about privacy, assurance, relationships with partner communities and support for emerging federations. In the summer 2010 REFEDs participants agreed to provide additional funding, which will allow REFEDs to progress work items more reliably, collaboratively and structurally and raise its visibility to related, external communities. This poster provides more details on REFEDs work-plan as well as the governance model.

  • Licia Florio - TERENA

Junos Space: programmable network automation platform

Juniper Networks¨ Junos¨ Space is an open, extensible network platform for developing and hosting applications that reduce cost and complexity and open your network to new service and revenue opportunities. It consists of: a scalable runtime environment to host applications; network infrastructure and operations management applications; hot-pluggable, multi-tenant environment to enable applications and devices to be dynamically integrated; a simple Web 2.0 user interface to speed productivity; and a complete rapid application development framework for developing simple and smart applications. Junos Space includes a portolio of collaborative Juniper-developed applications -- such as Route Insight, Ethernet Design, Service Now, and Virtual Control -- for scaling services, simplifying network operations, and automating support for complex network environments. One of the core members of the Junos platform, Junos Space uses the same design principles and open systems approach of the Junos operating system to enable programmability at the network application layer.

  • Debbie Montano - Juniper Networks

Fundamentals for Intelligent Application Networking on Campus

Networks are evolving from "just pipes" to being instrumental in the delivery of content and information for research, communications, and business applications. They must play a role in managing different classes of users (researchers, administration, students), prioritizing traffic types effectively to ensure end-to-end performance, and matching the connectivity available to diverse requirements effectively. The fundamental requirements to address Over the Top (OTT)content, multimedia communications, and traditional traffic (e-mail, etc.) types on today's networks focuses on the principles of Manage, Serve, Switch, and Connect. Learn more about Content Aware Networking, Layer 2-based flow control and SLA management, and how a high-bandwidth packet optical (Carrier Ethernet + WDM) underpinning provides the effective solution for campus and regional network applications. Key Technologies: Policy-based routing, Deep Packet Inspection, Transparent Content Caching, Carrier Ethernet, G.709 OTN, Packet Optical

  • Jason Smith - BTI Systems
  • Aleck Gilner - BTI Systems

High Performance Carrier Ethernet for Guaranteed Service Performance

FLASHWAVE Connection-oriented Ethernet (COE) combines the high performance, reliability, and security associated with Layer 1 based Ethernet over SONET/SDH (EoS) services with the flexibility and efficiency of native Ethernet. This solution allows service providers to retain the deterministic and secure advantages of traditional Layer 1 private-line services while incorporating the service-aware traffic management capabilities required to guarantee SLAs over a highly-scalable and reliable data architecture. The release 4.1 feature package enables high performance Carrier Ethernet networking opening up new business and wholesale opportunities not addressable with existing solutions. Next generation mobile backhaul, high definition video distribution and mission critical enterprise services require high performance, high availability connections only possible with COE. The frame delay, packet loss and service protection performance of non connection based networks is unacceptable for consistent delivery of critical voice, video and storage applications important to business and wholesale customers. Network provisioning and maintenance operations are simplified with NETSMART 1500 management integration allowing for a scalable deployment to thousands of COE services.

  • Craig Healey - Fujitsu Network Communications

Secure High-speed Transport for R&E Networks / Dynamically Reconfigurable Networks

ADVA Optical Networking will have 2 posters: 1. Secure High-speed Transport for R&E Networks - Data center / Cloud computing connectivity 2. Dynamically Reconfigurable Networks - Automated, flexible network resiliency

  • Brian Savory - ADVA Optical Networking


Provide information about eduroamus to the Internet2 membership.

  • Philippe Hanset - University of Tennessee Knoxville

Current status of Identity Management Federation in Japan

Japanese identify federation is now moved on the production level operation by means of the SAML 2.0 standard mainly utilizing Shibboleth middleware by Internet2. Major universities have already updated their status from pilot operation to production operation which utilizes the real account throughout the university. Through the practical operation, some issues and features were discussed and developed in order to enhance the federation. This presentation summarizes the progress and current status of our activity.

  • Kazu Yamaji - National Institute of Informatics
  • Motonori Nakamura - National Institute of Informatics

Analysis of the Datagram Congestion Control Protocol for Delay/Disruption Tolerant Networking

The Datagram Congestion Control Protocol (DCCP) was designed as a congestion controlled protocol for applications that care more about receiving data regularly than about retransmitting lost data. Such a protocol is ideal for VOIP, IPTV, and on-line games where you want congestion control, but don't need reliable delivery. This protocol offers several congestion control algorithms, including one that emulates TCP's congestion control and one that tries to reduce bandwidth fluctuations while still sharing the connection nicely. We layered the Licklider Transmission Protocol (LTP), a Delay/Disruption Tolerant Networking protocol designed for communication with endpoints in deep space, above DCCP in order to analyze LTP's reliable transmission capabilities across the terrestrial Internet as part of the dtnbone testing environment. In the process, we discovered several unexpected behaviors in the Linux DCCP implementation. As a result, we have developed several analysis tools and present an analysis of the Linux implementation's current behavior as an aid to fully understanding, and improving, its performance.

  • Samuel Jero - Ohio University

DING - A Protocol for Monitoring Delayed / Disrupted Networks

Delayed and disrupted networks pose an interesting problem when it comes to remote monitoring and / or management. Frequent data loss, transmission delays on the order of hours (or even days), and severe restrictions on available bandwidth combine to yield an environment that is often not ideal for existing protocols (e.g. SNMP). To this end, the Diagnostic Interplanetary Network Gateway (DING) protocol is being developed. This poster describes the structure and operation of the DING protocol. The publish / subscribe model DING uses is examined in detail. Further, the poster explains how DING attempts to overcome challenges inherent in a delayed / disrupted network. Additionally, DING's relationship with the bundle protocol (as described in RFC 5050) is highlighted. Finally, the poster illustrates a typical use-case for the DING protocol.

  • Gilbert Clark - Ohio University

Infinera's 100G PIC and its impact on the economics of future networks

The introduction of 100G PIC-based systems will have important implications for the economics of future networks. TomorrowÕs networks will require many more optical functions per wavelength than do todayÕs networks. This requirement stems from the advanced modulation formats needed for 100G transmission, and can be expected to lead to increased network cost and complexity if implemented using conventional optical components. To solve this problem, InfineraÕs 500G PICs incorporate more than 600 optical elements on a pair of indium phosphide chips, delivering significant benefits to next-generation Internet-centric networks, providing an effective means for service providers to scale network capacity while simultaneously obtaining lower space consumption, lower power consumption, increased reliability, and superior network economics. In the longer term, without photonic integration, ever-increasing bandwidth demand would continue to drive up network complexity, making photonic integration key to achieving superior carrier economics.

  • John Walker - Infinera

Indo-US Cancer Research Grid: A global infrastructure for biomedical informatics

The US National Cancer Institute (NCI) created the cancer Biomedical Informatics Grid (caBIG) program in 2004 to support personalized medicine (i.e. basing a patient's treatment on the specific biological nature of their disease) by creating a virtual network, or Grid, of interconnected people, institutions and data. CaBIG now supports a community of more than 1000 participants, with more than 150 interoperable information systems connected to its underlying infrastructure, caGrid, and numerous other systems in use locally. CaBIG is partnering with a variety of organizations in India including the Center for the Development of Advanced Computing (CDAC) to create a joint Indo-US Cancer Research Grid that will enable collaborations between US and Indian biomedical researchers as well as drive the next generation of interoperable infrastructure that will be available for use in both countries.

  • George A. Komatsoulis - National Cancer Institute
  • Anil Srivastava - National Cancer Institute