Home | Program | Keynotes | Tutorials | Videos | Committees | Sponsors | Previous Conferences | Contact
PLATINUM PATRONS
GOLD PATRONS
MEDIA PATRONS

Tutorial 1.1

Title

OpenFlow from a Developer's Perspective

Speakers

Srini Seetharaman, Deutsche Telekom R&D

Abstract

OpenFlow is an open interface for remotely controlling the forwarding tables in network switches, routers, and access points. Following the launch of the www.opennetworkingfoundation.org, OpenFlow is being widely embraced by the industry and GENI.

This tutorial is your opportunity to gain hands-on experience with the platform and programming tools most useful for developing network control applications on OpenFlow. Following an introduction to OpenFlow and the Open Programmable Extensible Networking (OPEN) architecture, each participant will create a flow-based Ethernet switch within a slice. Along the way, you'll learn the OpenFlow software suite: you'll view flow tables with dpctl, dissect packets with Wireshark, writing a NOX-based controller, simulate a multi-switch/multi-host network with Mininet (emulation environment) on your laptop, and slice it using the FlowVisor. Time permitting, we would include the use of Expedient/Opt-in Manager (GENI integration software) and other controller platforms.

Detailed Outline

  1. Lecture (45 min)
    1. Introduction: Why, What, How
    2. OpenFlow Potential, Limitations, Current vendors
    3. Big Picture: Software-defined networking
  2. Hands-on Portion (1 hour)
    1. Learn dpctl, wireshark and other tools
    2. self-exploration of mininet environment
    3. Build hub controller using NOX
    4. Build learning switch over NOX
  3. Lecture (30 min)
    1. Intro to Slicing, Virtualization and FlowVisor
    2. Details about GENI deployments
  4. Hands-on Portion (30 min)
    1. Review of NOX modules
    2. Slice using FlowVisor

Bios

Srini Seetharaman is a member of the Clean Slate Lab at Stanford University and a Senior Research Scientist with Deutsche Telekom R&D Lab, Los Altos, CA. He leads the OpenFlow deployment activities in the US as part of the GENI initiative. He is a recipient of Future Internet Design award from National Science Foundation. He holds a Ph.D. in Computer Science from the Georgia Institute of Technology and a Masters degree in Computer Science from The Ohio State University. His research interests include networking architectures and protocols, overlay networks, network monitoring and green technologies.


Tutorial 1.2

Title

Interconnection Networks for Cloud Data Centers

Speaker

Sudipta Sengupta, Microsoft Research

Abstract

Large scale data centers are enabling the new era of Internet cloud computing. The computing platform in such data centers consists of low-cost commodity servers that, in large numbers and with software support, match the performance and reliability of expensive enterprise-class servers of yesterday, at a fraction of the cost. The network interconnect within the data center, however, has not seen the same scale of commoditization or dropping price points. Today's data centers use expensive enterprise-class networking equipment and associated best-practices that were not designed for the requirements of Internet-scale data center services -- they severely limit server-to-server network capacity, create fragmented pools of servers that do not allow any service to run on any server, and have poor reliability and utilization. The commoditization and redesign of data center networks to meet cloud computing requirements is the next frontier of innovation in the data center.

Recent research in data center networks addresses many of these aspects involving both scale and commoditization. By creating large flat Layer 2 networks, data centers can provide the view of a flat unfragmented pool of servers to hosted services. By using traffic engineering methods (based on both oblivious and adaptive routing techniques) on specialized network topologies, the data center network can handle arbitrary and rapidly changing communication patterns between servers. By making data centers modular for incremental growth, the up-front investment in infrastructure can be reduced, thus increasing their economic feasibility. This is an exciting time to work in the data center networking area, as the industry is on the cusp of big changes, driven by the need to run Internet-scale services, enabled by the availability of low-cost commodity switches/routers, and fostered by creative and novel architectural innovations.

We will begin with an introduction to data centers for Internet/cloud services. We will survey several next-generation data center network designs that meet the criteria of allowing any service to run on any server in a flat un-fragmented pool of servers and providing bandwidth guarantees for arbitrary communication patterns among servers (limited only by server line card rates). These span efforts from academia and industry research labs, including VL2, Portland, SEATTLE, Hedera, and BCube, and ongoing standardization activities like IEEE Data Center Ethernet (DCE) and IEEE TRILL. We will also cover other emerging aspects of data center networking like energy proportionality for greener data center networks.

Detailed Outline

  1. Introduction to Cloud Data Centers (20 min)
    • What does a Cloud Data Center Look Like?
    • Data Center Costs, Complexity, and Scaling Requirements
    • Cisco Reference Network Design and Limitations
    • Networking Equipment: Traditional, Emerging, Commoditization Trends
  2. Clos Networks and Traffic Oblivious Routing (35 min)
    • Traffic Variation in MapReduce Data Centers
    • Traffic Oblivious Routing using Valiant Load Balancing
    • Implementation Aspects
    • Network Cost Comparisons
  3. Flat Layer 2 Network Design (45 min)
    • SEATTLE: Distributed Directory Service based on Network Layer One-hop DHT
    • VL2: End-Host Modifications and Centralized Directory Server Design
    • Portland: Multi-rooted Tree Networks, Positional Pseudo MAC
  4. Adaptive Routing (20 min)
    • Limitations of static hash-based multi-path routing (e.g., ECMP)
    • Hedera Architecture
    • Dynamic Flow Demand Estimation
    • Flow Scheduling Algorithms
    • Implementation Aspects
  5. Modular Data Center Network Design (30 min)
    • Trend towards Data Center Modularization
    • Modular Layouts of VL2 for Shipping Containers
    • BCube: Intra-container Data Center Network
  6. Energy Efficiency in Data Center Networks (30 min)
    • Energy Efficient Flattened Butterfly Topology
    • Dynamic Power/Performance Adjustment in Plesiochronous Links
    • Network-wide Power management using ElasticTree

Bio

Dr. Sudipta Sengupta is currently at Microsoft Research, where he is working on data center systems and networking, peer-to-peer applications, mobile connectivity, non-volatile memory for cloud/server applications, and data deduplication. Previously, he spent five years at Bell Laboratories, Lucent Technologies, where he advanced the state-of-the-art in Internet routing, optical switching, network security, wireless networks, and network coding.

Dr. Sengupta has taught advanced courses/tutorials on networking at many academic/research and industry conferences (please see list below). He received a Ph.D. and an M.S. in Electrical Engg. & Computer Science from Massachusetts Institute of Technology (MIT), USA, and a B.Tech. in Computer Science & Engg. from Indian Institute of Technology (IIT), Kanpur, India. He was awarded the President of India Gold Medal at IIT-Kanpur for graduating at the top of his class across all disciplines. He has published 65+ research papers in some of the top conferences, journals, and technical magazines, including ACM SIGCOMM, ACM SIGMETRICS, USENIX ATC, IEEE INFOCOM, IEEE International Conference on Network Protocols (ICNP), ACM SIGCOMM Internet Measurement Conference (IMC), International Conference on Distributed Computing Systems (ICDCS), Allerton Conference on Communication, Control, and Computing, Conference on Information Sciences and Systems (CISS), IEEE International Symposium on Information Theory (ISIT), ACM Hot Topics in Networking, International Conference on Very Large Data Bases (VLDB), ACM SIGMOD, IEEE/ACM Transactions on Networking (ToN), IEEE Journal on Selected Areas in Communications (JSAC), IEEE Transactions on Information Theory (ToIT), IEEE Communications Magazine, IEEE Network Magazine, ACM Symposium on Theory of Computing (STOC), European Symposium on Algorithms (ESA), Discrete Optimization, and Journal of Algorithms. He has authored 40+ patents (granted or pending) in the area of computer networking.

Dr. Sengupta won the IEEE Communications Society William R. Bennett Prize for 2011 and the IEEE Communications Society Leonard G. Abraham Prize for 2008 for his work on oblivious routing of Internet traffic. At Bell Labs, he received the President's Teamwork Achievement Award for technology transfer of research into Lucent products. His work on peer-to-peer based distribution of real-time layered video received the IEEE ICME 2009 Best Paper Award. At Microsoft, he received the Gold Star Award which recognizes excellence in leadership and contributions for Microsoft's long term success. Dr. Sengupta is a Senior Member of IEEE.


Tutorial 2.1

Title

Designing Scientific and Enterprise Computing Systems with InfiniBand and High-Speed Ethernet: Current Status and Trends

Speaker

D. K. Panda (The Ohio State University)

Abstract

InfiniBand (IB) and High-Speed Ethernet (HSE) interconnects are generating a lot of excitement towards building next generation scientific and enterprise computing systems. This tutorial will provide an overview of these emerging interconnects, the features they offer, their current market standing, and their suitability for designing these systems. It will start with a brief overview of IB, HSE and their architectural features. An overview of the emerging OpenFabrics stack which encapsulates both IB and Ethernet in a unified manner, and hardware technologies such as RDMA over Converged Enhanced Ethernet (RoCE) that aim at converged hardware solutions will be presented. IB and HSE hardware/software solutions and the market trends will be highlighted. Finally, sample performance numbers highlighting the performance these technologies can achieve in different environments such as MPI, PGAS/UPC, Parallel File Systems, Memcached and Hadoop will be shown.

Description

This half-day tutorial will bring current status and trends of different network interconnects (with emphasis on IB and HSE) to the audience in a single coherent presentation, focus on their individual strengths and limitations and provide a comparative study between these standards.

Based on these emerging trends and the associated challenges, the goals of this tutorial are as follows:
  • Making the attendees familiar with the IB and HSE network interconnects and the associated benefits.
  • Providing a comparative study between IB and HSE.
  • Demonstrating how the OpenFabrics stack is trying to provide a convergence between these two standards.
  • Providing an overview of available IB and HSE hardware/software solutions and trends
  • Outlining case studies of designing next generation systems (HPC with MPI, PGAS/UPC, File Systems and emerging enterprise datacenters with Memcached and Hadoop) while taking advantage of IB and HSE features and multi-core computing platforms.
This tutorial is targeted for various categories of people working in the areas of high performance communication and I/O, storage, networking, middleware, virtualization, and applications related to scientific computing and enterprise computing systems:
  • Newcomers to the field of high-speed networking who want to familiarize themselves with IB and HSE technologies.
  • Scientists, engineers, and researchers working on the design and development of next generation high-end computing systems including computing clusters, data centers and storage centers.
  • Developers of next generation networked computing middleware and applications.
  • Managers and administrators responsible for setting-up next generation scientific/enterprise computing systsms and facilities in their organizations/laboratories.
There is no fixed pre-requisite. As long as the attendee has general knowledge in high performance computing, net- working, storage, and related issues, he/she will be able to understand and appreciate it. The tutorial is designed in such a way that an attendee gets exposed to the topics in a smooth and progressive manner.

Outline

The tutorial is organized along the following topics:
  1. What are IB and HSE?
    • TCP vs. User-level communication protocols
    • Requirements (communication, I/O, performance, cost, RAS) from the perspective of designing next generation scientific and enterprise computing systems
  2. Short Overview of InfiniBand Architecture
    • Architecture and Basic Hardware Components
    • Novel Features (Hardware Protocol Offload and Communication Semantics)
    • IB Verbs Interface
    • Management and Services (Subnet Management and Hardware Support for Scalable Network Management)
    • Overview of different IB protocols (Verbs, IPoIB and SDP)
  3. Overview of High Speed Ethernet, Convergence and Features
    • Architecture and Basic Components (Similarity and Differences with InfiniBand)
    • RDMA over Converged Ethernet (RoCE)
    • Software and Hardware based iWARP on 10GigE
    • Existing Implementations of HSE
  4. Overview of IB and HSE Products (hardware and software), Time-frames, and Market Trends
    • Vendors, Switches, and Host Channel Adapters
    • Overview of ConnectX architecture
    • Overview of OpenFabrics Architecture and Convergence
    • Pointers to IB and HSE installations
  5. Designing High-end Systems with IB and HSE: Research Challenges, Case Studies and Performance Evaluation
    • High-end clusters with MPI-1 and MPI-2
    • Partitioned Global Address Space (PGAS/UPC) Design
    • Parallel File Systems
    • Enterprise Computing Systems (Hadoop and Memcached)
    • InfiniBand in the Cloud (Virtualization and IB-WAN)
  6. Conclusions, Final Q&A, and Discussion

Bio

Prof. D. K. Panda: Dhabaleswar K. (DK) Panda is a Professor of Computer Science at the Ohio State University. He obtained his Ph.D. in computer engineering from the University of Southern California. His research interests include parallel computer architecture, high performance computing, communication protocols, files systems, network-based com- puting, and Quality of Service. He has published over 270 papers in major journals and international conferences related to these research areas. Dr. Panda and his research group members have been doing extensive research on modern networking technologies including InfiniBand, 10GE/iWARP and RDMA over Ethernet (RoCE). His research group is currently collab- orating with National Laboratories and leading InfiniBand and 10GE/iWARP companies on designing various subsystems of next generation high-end systems. The MVAPICH/MVAPICH2 (High Performance MPI over InfiniBand, iWARP and RD- MAoE) open-source software packages, developed by his research group, are currently being used by more than 1,175 organizations worldwide (in 59 countries). This software has enabled several InfiniBand clusters (including the 6th, 7th and 11th ranked ones) to get into the latest TOP500 ranking. These software packages are also available with the Open Fabrics stack for network vendors (InfiniBand, iWARP and RoCE), server vendors and Linux distributors. Dr. Panda's research is supported by funding from US National Science Foundation, US Department of Energy, and several industry including Intel, Cisco, SUN, Mellanox, QLogic and NetApp. He is an IEEE Fellow and a member of ACM. More details about Prof. Panda are available here.


Tutorial 2.2

Title

The Evolution of Network Architecture towards Cloud-Centric Applications.

Speaker

Loukas Paraschis, Cisco

Abstract

The increasing availability of fast and reliable Internet access has enabled applications to transition to an Internet based service delivery model, commonly referred to as "cloud computing". The underlying infrastructure consists of data-centers of massive compute and storage resources, and networking which is crucial in interconnecting and optimizing the costperformance of the "cloud infrastructure". As a result the interconnection of data-centers is one of the largest contributors to the increase of traffic demand in the traditional backbone transport networks, and consequently the structure of the Internet has started to change towards a flatter hierarchy and a denser interconnection with consequences for routing, traffic engineering and security.

We review the unique network requirements, design challenges, and desired future hardware and software features of these systems and their components. We first analyze the functional characteristics and challenges of these networks, and review the current and emerging applications that motivated these networks to scale leveraging IP, MPLS, and DWDM transport. We particularly discuss how the new, high-bandwidth, predominantly video related, applications (including IPTV, video-on-demand, peer-to-peer, and video-conferencing), often with diverse quality-of-service requirements, are increasingly motivating a fundamental shift in services from circuits to packets, giving rise to the most significant evolution of transport networks in recent history. The tutorial then focuses on the current and future converged WDM transport. We review the significant advancements in optical technologies, system, standards, and networking architectures, and the improvement in capital and operational cost, including bandwidth, density, or power. We finally attempt evaluate the interplay among the intra and inter data-center networking architectures, system design, and the enabling photonics technology and packaging innovations. Future network evolution, emerging standards, and related research topics are also being considered.

Outline

Converged IP and WDM network architectures have increasingly become the best answer addressing the transport network needs. This tutorial reviews the current and emerging "cloud-centric" network developments, and evaluates the functionality, characteristics, and the associated challenges, along with the interplay among important and promising photonics technologies.

A summary of content and time allocation is:
  • Network Architecture Review, Key Applications: 30-60 minutes
  • Evolution, & Current Challenges: 30-60 minutes
  • Current Technologies, and State-of-the-art system design: 60-120 minutes
  • Emerging technologies, Innovation, and Trends: 15-45 minutes

This tutorial/course will enhance the audience understanding of the interplay between network architectures, systems, and photonics technology innovation in the actual evolution (past, current and future) of the public network infrastructure.

This tutorial/course is intended for researchers (and students) in the field of photonics networking that wish to obtain an industry perspective, and also industry professionals that wish to have a network architecture and system level analysis of the networking evolution, with a particular focus on photonics related technologies.

Bio

Loukas (Lucas) Paraschis is solutions business development manager at cisco, responsible for next generation core network architectures in emerging markets. At cisco, he has worked also on IPover-WDM architectures, multi-service metro WDM systems, and optical transport technologies, and the associated market development efforts. Prior to his current role, Loukas worked as an R&D engineer, product manager, and technical leader in optical networking and core routing, and completed graduate studies at Stanford University (PhD applied physics 1999, MS EE 1998). He has (co)authored more than 70 peer-reviewed publications, invited, and tutorial presentations, a book, a book chapter, technical reports, and three patents, has been associate editor for the Journal of Optical Communication and Networks, guest editor of the IEEE Journal of Lightwave Technology, member of the IEEE (SM'06), the OSA, and multiple conference organizing committees, and is an IEEE Photonics Society Distinguished Lecturer (2009). Loukas was born in Athens, Greece, where he completed his undergraduate studies.

SILVER PATRONS