14 September 2009, Issy-les-Moulineaux, France

The workshop program is available here.

Introduction to DC CAVES
Several studies over the past few years have found that enterprises are spending too much to manage the operational aspects of the complex application environments they have deployed over time. The recent economic downturn is placing additional requirements to lower the capital expenses, while delivering more computing, networking and storage capabilities. These operational and capital demands are causing a significant shift in Data Center (DC) infrastructure, from a traditional compute model, which has lightly utilized servers using 1 Gbps network adapters to a new virtual and converged compute model, which has highly utilized servers using 10 Gbps Converged Network Adapters (CNAs) with many Virtual Servers that attach to converged fabrics (iSCSI, Network Attached Storage - NAS, or Fibre Channel over Ethernet - FCoE).

Network convergence and virtualization technologies associated with this new model extend the IT skill set, which, if harnessed, can allow the enterprise to reap significant savings. For networking within the Data Center the potential value comes from: lowering capital expenses through higher utilization (server, storage and network), and converged fabrics; and lowering operational expenses through automated and integrated management that optimizes Data Center infrastructure. At the edge of the Data Center (DC), virtual networking (e.g. MPLS, VPLS) as well as new virtualization paradigms also offer tremendous savings associated with combining networks, especially across wide-area networks interconnected with costly WAN links.

Scope
Organized together with ITC 21, the first DC CAVES workshop is intended to serve as a forum to present the latest work by researchers and developers from both academia and industry. The workshop focuses on the technologies that will be needed to meet the demands of the new virtual and converged compute model described above. For each of these technologies the workshop will analyze the problem and solution alternatives.

The two major topics of interest are server virtualization infrastructure and physical switch virtualization infrastructure. Server virtualization infrastructure includes: layer-2/3/ switching technologies performed within the server, such as: Hypervisor Virtual Ethernet Switches, Server Input/Output Virtualization and Virtual Appliances running in Server Virtual Machines. Physical switch virtualization and convergence infrastructure includes: layer-2/3/+ fabric virtualization technologies, such as VLANs, MPLS, VPLS, Switch Stacking and mechanisms that partition a single physical switch into multiple virtual switches; and storage and Ethernet convergence technologies, such as iSCSI, NAS, FCoE and FC over MPLS. Additional areas of interest for this workshop are:

•  Server virtualization infrastructure
  • Automation of virtual server network identity management
  • Enhanced virtual server network access and traffic controls
  • Networking technologies to enable server migration within an entire DC
  • Networking technologies to enable virtual server migration across DCs
  • Security plane infrastructure virtualization
  • Enhancements to switches used by virtualization intermediaries (e.g. Hypervisors)
  • Offloading of virtual switching to external fabrics
  • Converged fabric reference services architectures
  • Future directions

•  Virtual & converged fabric infrastructure
  • Overall network virtualization & performance
  • Performance evaluation of converged iSCSI, NAS and emerging FCoE fabrics
  • Converged fabric security considerations
  • Transport stack options converging Inter-Process Communication (IPC) traffic
  • Additional Ethernet Quality of Service enhancements needed for converged environments
  • Performance and fault event management for converged fabrics
  • Converged fabric management infrastructure
  • Converged fabric reference services architectures
  • Future directions

Workshop Program Chair

• Renato Recio, IBM, USA

Committee Members

• Thierry Coupaye, Orange Labs, France
• Uri Elzur, Broadcom Corporation, USA
• Claus Gruber, Detecon Consulting, Germany
• Michael Kagan, Mellanox Technologies, Israel
• David Kahn, Sun Microsystems, USA
• Andreas Kirstaedter, University of Stuttgart, Germany
• Mike Krause, Hewlett-Packard Company, USA
• Marco Hoffmann, Nokia Siemens Networks, Germany
• Mallik Mahalingam, VMware, USA
• Michael Menth, University of Wuerzburg, Germany
• Aki Nakao, University of Tokyo, Japan
• Dhabaleswar Panda, Ohio State University, USA
• Joe Pelissier, Cisco Systems, USA
• Ashley Saulsbury, Juniper, USA
• Kurt Tutschku, University of Vienna, Austria
• Manoj Wadekar, QLogic, USA
• Suresh Vobbilisetty, Brocade, USA

Workshop Program

Top