190 likes | 315 Views
SAHARA Second Winter Retreat 13-15 January 2003. Randy H. Katz, Anthony Joseph, Ion Stoica Computer Science Division Electrical Engineering and Computer Science Department University of California, Berkeley Berkeley, CA 94720-1776. UC Berkeley Project Team. Industrial Collaborators Friends.
E N D
SAHARA Second Winter Retreat13-15 January 2003 Randy H. Katz, Anthony Joseph, Ion Stoica Computer Science Division Electrical Engineering and Computer Science Department University of California, Berkeley Berkeley, CA 94720-1776
UC Berkeley Project Team Industrial Collaborators Friends Retreat Goals &Technology Transfer People Project Status Work in Progress Prototype Technology Early Access to Technology Promising Directions Reality Check Feedback
AT&T Research Yatin Chawathe Cisco Silvan Gai David Jaffe Crazy Tulip Systems Chris Overton Ericsson Research Yuri Ismailov Hewlett-Packard Labs Wai-Tian Dan Tan Susie Wee Intel Research -Timothy Roscoe (ROC) Microsoft Research Lili Qui Helen Wang NEC Yasuhiko Matsunaga (VIF) Nortel Networks Tal Lavian (PhD student) Håkan Millroth NTTDoCoMo Takashi Suzuki (VIF) Gang Wu Rhapsody Networks Brian Byun Sprint ATL Paul Jardetzky Univ. Helsinki Kimmo Raatikainen (+ Nokia Research) Univ. NSW Aruna Seneviratne Other Affiliation Bryan Lyles Who is Here (Industry) Italics indicates Ph.D. from Berkeley VIF=Visiting Industrial Fellow Green = First Retreat!
Professors Anthony Joseph Randy Katz Ion Stoica Doug Tygar Postdocs Kevin Lai Technical & Admin Staff Bob Miller Keith Sklower Grad Students Dan Adkins Sharad Agarwal Matt Caesar Weidong Cui Steve Czerwinski Grad Students Ling Huang Karthik Lakshminarayanan Yin Li Anshi Liang Huang Ling Sridhar Machiraju George Porter Anantha Rajagoplala-Rao Lakshmi Subramanian Mel Tsai Fang Yu Shelley Zhuang Ana Sanchez Merino(on leave from Ericsson) Who is Here (Berkeley)
Retreat Purpose • Third SAHARA Retreat • Project launched 1 July 2001 • Halfway: review progress, set directions, consider “next” project • Goal: Explore architectural elements for future networks • “Services” inside the network: code vs. protocols, location/topology-aware • Spanning: • Independent service providers • Converged data + telecomms nets • Hetero access + core nets • Leverage Co-lo w/ ROC Retreat • Reliable Computing + Comms + Services = New Gen Distributed Applications • Industrial feedback & directions • Real-world networking problems/limitations • Helping us do relevant research at Internet-scale
“Elevator” Statement • New mechanisms, techniques for end-to-end services w/ desirable, predictable, enforceable properties spanning potentially distrusting service providers • Architecture for service composition and inter-operation across separate administrative domains, supporting peering and brokering, and diverse business, value-exchange, access-control models
Service Composition Layered Reference Model for Service Composition End-User Applications Applications Services Application Plane Middleware Services End-to-End Network With Desirable Properties Enhanced “Paths” Connectivity Plane Enhanced “Links” IP Network
Project Status: Top Down • Initial Emphasis on Application Plane • Distributed Telecom-oriented Applications: messaging, content distribution, voice • Universal In-box, CDNs, VoIP, Broadcast/Multicast • Shelley Zhuang M.S. Project (12/01), “Bayeux: An Architecture for Scalable and Fault-tolerant Wide-area Data Dissemination” • Mukund Seshadri, M.S. Project (12/02), “A Scalable Architecture for Broadcast Federation” • + work by Jimmy (VoIP), Matt (VoIP),Morley (CDN), Yan (CDN) • Service Composition • Morley Mao’s M.S. Project (12/00), “Fault-tolerant, Scalable, Wide-Area Internet Service Composition” • Helen Wang’s Ph.D. Dissertation (12/01): “Scalable Robust Wide-Area Architecture for Unified Communications” • Bhaskar Raman’s Ph.D. Dissertation (12/02): “An Architecture for Availability and Performance in Wide-Area Service Composition” • Resource Management • Chee-Nee Chuiah’s Ph.D. Dissertation (12/01), “A Scalable Framework for IP-Network Resource Provisioning Through Aggregation and Hierarchical Control” • Jimmy Shih’s Ph.D., “Congestion Pricing for Network Resource Allocation”
Summer Retreat Feedback • Focus on synergistic research activities: • Resilence, trust, failure recovery of routing infrastructure • Interdomain routing: BGP, Verification, Policy Layer • Overlay networks: evolution to new protocols/services • Consider the following: • Enterprise vs. ISP viewpoints: provisioning, monitoring • Integration with mobility and access networks • Implications of streaming media • Resource management • Define criteria for correctness: learn from other successful robust distributed systems (e.g., DNS, Mail, News, etc.) • Real deployments (PlanetLab) • Involve more network equipment industry
Project Status: Bottom Up • Renewed emphasis on Connectivity Plane • “Reachability” as a Network Service • Implementing paths between composed service instances,e.g., “links” within an overlay network • Multi-provider environment, no centralized control • Evolve interdomain routing for desirable, controllable properties • Overlays networks as alternative to protocol evolution, focusing on new kinds of desirable properties • Trust: verify believability of routing advertisements • Agility: converge quickly in response to global routing changes to retain good reachability “performance” (e.g., latency)? • Reliability: detect service composition path failures quicklyto enable fast recomposition to maintain reachability • Scalability and Interoperability: Adapt protocols via processing at “impedance” matching points between administrative domains
New Directions • OASIS: Overlays and Active Services for Internetworked Storage • Wide-Area Network-attached Storage Services, particularly for disaster recovery • Composed Services and Resource Management • Authorization Control Across Administrative Domains (Suzuki, Moreno + students) • Radio Resource Allocation Across Service Providers (Matsunaga + students)
Routing as a Composed Service • Routing as a Reachability “Service” • Implementing paths between composed service instances,e.g., “links” within an overlay network • Multi-provider environment, no centralized control • Desirable Properties • Trust: verify believability of routing advertisements • Agility: converge quickly in response to global routing changes to retain good reachability “performance” (e.g., latency)? • Reliability: detect service composition path failures quicklyto enable fast recomposition to maintain reachability • Scalability and Interoperability: Adapt protocols via processing at “impedance” matching points between administrative domains
Convergence Issues Shared Bottlenecks New Control Plane New Gen Routing Infrastructure Overlay Routing QoS and Performance Failure Detection and Recovery Measurement-Based Algorithms Validity and Security Research Strategy Existing Interdomain Routing
Plan for the Retreat • Monday, 13 January 2003 • 0800-1200 Bus to Tahoe • 1200-1330 Lunch • 1330-1500 Retreat Overview and Introductions (Randy) • Retreat Overview & Sahara Progress, Randy • I3 Status, Ion • Griffin Status, Anthony • Tapestry/Oceanstore Intro, Anthony • 1500-1530 Break • 1530-1700 Routing (Ion) • Verifiable Routing (Lakshmi) • Interdomain Routing Control and Policies (Sharad) • Detecting Bottlenecks (Machi, Weidong) • 1700-1730 Short Break to Re-arrange Rooms • 1730-1830 Long-lived Distributed Systems (Kubi) • Project Seagull (Hakim) • Benchmarking for P2P Systems (David) • 1845-2000 Joint Dinner • 2000-2100 A New Research Agenda for Systems (Dave and Randy) • 2100- Student Posters & Social Hour
Plan for the Retreat • Tuesday, 14 January 2003 • 0730-0830 Breakfast • 0830-1000 OceanStore/I3 (Kubi) • Tapestry Deployment (Ben) • Results in Overlay Benchmarking (Sean) • Fault Tolerance/Locality in Tapestry (Jeremy/Kris) • Secure I3 (Dan) • 1000-1030 Break • 1030-1200 Overlay Routing (Ion) • Fast Failure Detection (Shelley, Matt) • Internet Iso-bar: A Scalable Overlay Distance Monitoring System (Yan) • Shared API for Overlay Networks (Ben) • 1200-1300 Lunch • 1300-1630 Long Break • 1630-1800 OASIS (Randy) • SWAN Overview (Randy) • Active Storage Networking Testbed (Mel) • Research Opportunity Discussion (Li, George) • 1800-1930 Dinner (Joint with ROC Retreat) • 1930-2100 Industry Wild Ideas & Open Mike (Armando) • 2100- Student Posters & Social Hour
Plan for the Retreat • Wednesday, 15 January 2003 • 0730-0830 Breakfast • 0830-1000 New Research Opportunity Synthesis (Randy & Dave) • 1000-1030 Break/Room Checkout/Photo Session • 1030-1200 Industrial Feedback (Randy) • 1200-1300 Lunch • 1300-1700 Bus back to Berkeley
Recent Publications • J. Shih, R. H. Katz, “Evaluating the Tradeoffs of Congestion Pricing for Voice Calls,” 2002 International Symposium on Performance Evaluation of Computer and Telecommunication Systems (SPECTS 2002), San Diego, California, (July 2002). • B. Raman, R. H. Katz, “Emulation-based Evaluation of an Architecture for Wide-Area Service Composition,” 2002 International Symposium on Performance Evaluation of Computer and Telecommunication Systems (SPECTS 2002), San Diego, California, (July 2002). • Z. Mao, R. Govindan, G. Varghese, R. H. Katz, “Route Flap Damping Exacerbates Internet Routing Convergence.” ACM SIGCOMM Conference, Pittsburgh, PA, (August 2002). • Y. Chen, R. H. Katz, J. D. Kubiatowicz, “SCAN: a Dynamic Scalable and Efficient Content Distribution Network,” International Conference on Pervasive Computing (Pervasive 2002), Zurich, Switzerland, (August 2002). • B. Raman, S. Agarwal, Y. Chen, M. Caesar, W. Cui, P. Johansson, K. Lai, T. Lavian, S. Machiraju, Z. Mao, G. Porter, T. Roscoe, M. Seshadri, J. Shih, K. Sklower, L. Subramanian, T. Suzuki, S. Zhuang, A. D. Joseph, R. H. Katz, I. Stoica, “The SAHARA Model for Service Composition Across Multiple Providers,” International Conference on Pervasive Computing (Pervasive 2002), Zurich, Switzerland, (August 2002), Invited Paper. • Z. Mao, R. H. Katz, “A Framework for Universal Service Access using Device Ensembles,” CRA Grace Murray Hopper Celebration of Women in Computer Science Conference, Vancouver, BC, (October 2002). Mao selected as Hopper Young Investigator (Best Student Paper).
Recent Publications • L. Subramanian, I. Stoica, H. Balakrishnan, R. H. Katz, “OverQoS: Offering QoS using Overlays,” First Workshop on Hot Topics in Networking (HotNets02), Princeton, NJ, (October 2002). • Y. Chen, L. Qui, R. H. Katz, “On the Clustering of Web Content for Efficient Replication,” 10th IEEE Conference on Network Protocols (ICNP 2002), Paris, France, (November 2002). • W. Cui, I. Stoica, R. H. Katz, “Backup Path Allocation Based on a Correlated Link Failure Probability Model in Overlay Networks,” 10th IEEE Conference on Network Protocols (ICNP 2002), Paris. France, (November 2002). • S. Agarwal, C. N. Chuah, R. H. Katz, “OPCA: Robust Interdomain Policy Routing and Traffic Control,” Proceedings OpenArch 2003, San Francisco, CA, (April 2003). • S. Zhuang, K. Lai, I. Stoica, R. H. Katz, S. Shenker, “Host Mobility using an Internet Indirection Infrastructure,” First International Conference on Mobile Systems, Applications, and Services (ACM/USENIX Mobisys), San Francisco, CA, (May 2003). • B. Raman, R. H. Katz, “Load Balancing and Stability Issues in Algorithms for Service Composition,” IEEE Infocomm Conference, San Francisco, California, (July 2003).
SaharaOverviewRandy H. KatzUniv. of CaliforniaBerkeley, CA94720-1776