Week Day
(leader)		 Technical Reading Homework/Extra reading
Data architectures
Jan 22M
(Mike)		 Http/caching:
Design Principles for Distributed Caching on the Internet Renu Tewari, Michael Dahlin, Harrick Vin, and John Kay. The 19th IEEE International Conference on Distributed Computing Systems (ICDCS). May 1999.
For more depth:
  • An Evaluation of the Ninth SOSP Submissions -or- How (and How Not) to Write a Good Systems Paper Roy Levin and David D. Redell, Ninth SOSP Program Committee Co-chairmen
    		•
    W
    (Mike)    		 Content distribution networks
    Fast Internet Content Delivery with FreeFlow (Akamai whitepaper)
    For more depth:     		Paper reviews are due for this and all subsequent lectures.
    Jan 29M
    (gold)    		 P2P
    Freenet: A Distributed Anonymous Information Storage and Retrieval System Clarke, Sandberg, Wiley, Hong
    For less depth: For what types of workload is a P2P architecture useful (and when would you be better off just using a big central server to, perhaps, "simulate" P2P functionality?     		The Task of the Referee Alan Jay Smith
    W
    (blue)    		 Streaming
    Measurement and Analysis of a Streaming Workload Maureen Chesire, Alec Wolman, Geoff Voelker, Henry Levy. USITS01.
    For more depth:     		Distribute HW1: Cache simulator
    Feb 5M
    (ruby)    		 Dynamic content
    "Active Naming: Flexible Location and Transport of Wide-area resources" A. Vahdat, T. Anderson, M. Dahlin pdf ps The 1999 USENIX Symposium on Internet Technologies and Systems (USITS99), October 1999
    For more depth: Should dynamic content be handled by shipping code to edges of network, or are "hooks" for special cases (caching dynamic results, hit count reporting, ...) enough?
    Sermon: Simplicity (Tom Anderson)
    Sermon: Performance Tuning (Tom Anderson)
    Begin cache simulator HW
    W
    (A)    		 Prefetch/push
    Prefetching Hyperlinks Dan Duchamp. USITS99.
    For more depth: Can latency be improved by proactively replicating data? By how much? And how aggressively should we replicate? With all of these prefetching algorithms, what is the design space for prefetching?
    Feb 12M
    (gold)    		 Rate of change/delta encoding
    Potential benefits of delta-encoding and data compression for HTTP, Jeffrey C. Mogul, Fred Douglis, Anja Feldmann, and Balachander Krishnamurthy
    For more depth: How much will data change limit replication? How much will similarity of versions help replication?
    W
    (blue)    		 Resource mgmt
    Resource management for scalable disconnected access to web servicesBharat Chandra, Mike Dahlin, Lei Gao, Amjad-Ali Khoja, Amol Nayate, Asim Razzaq, Anil Sewani. November 2000. To appear WWW10.
    For more depth:     		Cache simulator HW due.
    DEFER THIS DISCUSSIONM
    (Arun)    		 Data placement
    Bandwidth Constrained Placement in a WANVenkataramani, Weidmann, Dahlin
    For more depth:
    Sermon: Programming as Craft (Tom Anderson)
    Sermon: Stay Broad (Tom Anderson)
    Feb 21,26,28W/M
    (Mike)    		 Wrap up: Open questions (Wednesday class will be in ACES 6.442 as will Monday 2/26's class; Monday 2/26's class will begin at 4:30). Hw: open question
    Reliability/Availablity
    Mar 5W
    (Mike)    		 Routing pathologies
    End-to-End Routing Behavior in the Internet.V. Paxson. IEEE/ACM Transactions on Networking, Vol.5, No.5, pp. 601-615, October 1997
    For more depth:     		LESS SW engineering manual (read section 1 only)
    March 7W
    (A)    		 Routing pathologies
    Experimental Study of Internet Stability and Wide-Area Backbone Failures C. Labovitz, A. Ahuja, F. Jahanian. FTCS99.
    For more depth:
    March 12SPRING BREAK
    March 19M
    (gold)    		 Service Availability
    End-to-end WAN Service AvailabilityB. Chandra, M. Dahlin, L. Gao, A. Nayate. USITS 2001.
    For more depth:     		HW: Begin NS Tutorial [Lampson 83] B. W. Lampson. Hints for Computer System Design. Proc. of the 9th ACM Symposium on Operating Systems Principles, October 1983, pp. 33-48.
    Mar 21W
    (Mike)    		 Reliability wrap up ACES 6.112
    Do brainstorming from previous reliability discussions.
    Hot topics (USITS 2001 and others)
    March 26M
    		 Zorn talk (two weeks ago; take today off...)
    W
    Arun, Sadia, Praveeen, Ravi    		 Class project results: Potential costs and benefits of long-term prefetching for content-distribution
    Apr 2M
    (blue)    		 An Architecture for Content Routing Support in the Internet Mark Gritter and David R. Cheriton, USITS01 HW: NS Tutorial Due
    W
    (ruby)    		 On-line Markets for Distributed OBject Services: The MAJIC System, Levy, Blumrosen, Nisam. USITS01. For more depth:
    SkipM
    (A)    		 Puppeteer: Component-based Adaptation for Mobile Computing Eyal de Lara, Dan S. Wallach, and Willy Zwaenepoel, USITS01
    W
    (gold)    		 Measuring Client-Perceived Response Time on the WWW Ramakrishnan Rajamony and Mootaz Elnozahy
    Nettimer: A Tool for Measuring Bottleneck Link Bandwidth Kevin Lai and Mary Baker, Stanford University USITS01
    SkipM
    (blue)    		 Neptune: Scalable Replication Management and Programming Support for Cluster-based Network Services Kai Shen, Tao Yang, Lingkun Chu, JoAnne L. Holliday, Douglas A. Kuschner, and Huican Zhu, University of California, Santa Barbara USITS01
    W
    (ruby)    		 CSP: A Novel System Architecture for Scalable Internet and Communication Services Hemal V. Shah, Dave B. Minturn, Annie Foong, Gary L. McAlpine, Rajesh S. Madukkarumukumana, and Greg J. Regnier, Intel Corporation USITS01
    Network issues
    Apr 9M
    (Mike)
    NW v. web
    Network Performance Effects of HTTP/1.1, CSS1, and PNG, Henrik Frystyk Nielsen, Jim Gettys, Anselm Baird-Smith, Eric Prud'hommeaux, Håkon Lie, Chris Lilley SIGCOMM97
    For more depth:     		C. A. R. Hoare. The Emperor's Old Clothes. Communications of the ACM, Vo. 24, No. 2, February 1981, pp. 75-83.
    W
    (A)    		 Performance prediction/mapping
    On Estimating End-to-End Network Path Properties, Mark Allman SIGCOMM99
    For more depth
    • Measuring Client-Perceived Response Time on the WWW Ramakrishnan Rajamony and Mootaz Elnozahy USITS01
    • Nettimer: A Tool for Measuring Bottleneck Link Bandwidth Kevin Lai and Mary Baker, Stanford University USITS01
    Apr 16M
    (gold)    		 Performance prediction/mapping
    Using pathchar to Estimate Internet Link Characteristics Allen B. Downey, SIGCOMM99
    For more depth:
    W
    (blue)    		 Mapping/Anycast
    On the placement of Internet Instrumentations Proc of IEEE Infocom 2000, March 2000, S. Jamin, C. Jin, Y. Jin, D. Raz, Y. Shavitt, and L. Zhang
    For more depth:
    Apr 23M
    (ruby/A)    		 Flat pricing v. prioritized traffic
    (ruby:) Best-Effort versus Reservations: A Simple Comparative Analysis Lee Breslau and Scott Shenker SIGCOMM98
    (A:) Should flat-rate Internet pricing continue?Andrew Odlyzko Computer Networks
    For more depth:
    Apr 24Tu
    ACES 2.302 11AM    		 Guest lecture: Helen Wang "Scalable, Robust Widea-area control architecture for Integrated Communications" Read the ICEBERG paper
    Apr 25W
    (ruby/A)    		 continue flat v. pricing debate
    Also, how to improve reliability?
    For more depth
    Apr 30M Guest lecture

    For more depth:
    		Turn in group discussion peer review.
    W Class rescheduled to Apr 24 11AM