Mining Web Traces: Workload Characterization, Performance Diagnosis, and Applications

1
Mining Web TracesWorkload Characterization,
Performance Diagnosis, and Applications

• Lili Qiu
• Microsoft Research
• Internal Talk
• November 2002

2
Motivation

• Why do we care about Web traces?
• Content providers
• How do users come to visit the Web site?
• Why do users leave the Web site? Is poor
  performance the cause for this?
• Where are the performance bottlenecks?
• What content are users interested in?
• How does users interest change in time?
• How does users interest change across different
  geographical regions?

3
Motivation (Cont.)

• Web hosting companies
• Accounting billing
• Server selection
• Provisioning server farms where to place servers
• ISPs
• How to save bandwidth by storing proxy caches?
• Traffic engineering provisioning
• System designers
• Where are the performance bottlenecks?
• How to improve Web performance?
• Examples Traffic measurements have influenced
  the design of HTTP (e.g., persistent connections
  and pipeline), TCP (e.g., initial congestion
  window)

4
Outline

• Background
• Web workload characterization
• Performance diagnosis
• Applications of traces
• Bibliography

5
Part I Background

• Web software components
• Web semantic components
• Web protocols
• Types of Web traces

6
Web Software Components

• Web clients
• An application that establishes connections to
  send Web requests
• E.g., Mosaic, Netscape Navigator, Microsoft IE
• Web servers
• An application that accepts connections to
  service requests by sending back responses
• E.g., Apache, Microsoft IIS
• Web proxies (optional)
• Web replicas (optional)

Internet
proxy
replica
proxy
replica
proxy
WebServers
WebClients
7
Web Semantic Components

• Uniform Resource Identifier (URI)
• An identifier for a Web resource
• Name of protocol http, https, ftp, ..
• Name of the server
• Name of the resource on the server
• e.g., http//www.foobar.com/info.html
• Hypertext Markup Language (HTML)
• Platform-independent styles (indicated by markup
  tags) that define the various components of a Web
  document
• Hypertext Transfer Protocol (HTTP)
• Define the syntax and semantics of messages
  exchanged between Web software components

8
Example of a Web Transaction
DNS server
1. DNS query
2. Setup TCP connection
3. HTTP request
4. HTTP response
Browser
Web server
9
Internet Protocol Stack
Application layer application programs (HTTP,
Telnet, FTP, DNS)
Transport layer error control flow control
(TCP,UDP)
Network layer routing (IP)
Datalink layer handle hardware
details (Ethernet, ATM)
Physical layer moving bits (coaxial cable,
optical fiber)
10
HTTP Protocol

• Hypertext Transfer Protocol (HTTP)
• HTTP 1.0 BLFF96
• The most widely used HTTP version
• A Stop and wait protocol
• HTTP 1.1 GMF99
• Persistent connections use one TCP connection
  for multiple HTTP requests
• Pipelining send multiple requests without
  waiting for a response between requests
• Content negotiation
• Range requests
• Caching control

11
Types of Web Traces

• Application level traces
• Server logs CLF and ECLF formats
• CLF formatltremoteIP,remoteID,usrName,Time,request
  ,responseCode,contentLengthgte.g., 192.1.1.1, -,
  -, 8/1/2000, 100000, GET /news/index.asp
  HTTP/1.1, 200, 3410
• Proxies logs CLF and ECLF formats
• Client logs no standard logging formats
• Packet level traces
• Collection method monitor a network link
• Available tools tcpdump, libpcap, netmon
• Concerns packet dropping, timestamp accuracy

12
Tutorial Outline

• Background
• Web workload characterization
• Performance diagnosis
• Applications of traces
• Bibliography

13
Part II Web Workload Characterization

• Overview of workload characterization
• Content dynamics
• Access dynamics
• Common pitfalls
• Case studies

14
Overview of Workload Characterization

• Process of trace analyses
• Common analysis techniques
• Common analysis tools
• Challenges in workload characterization

15
Process of Trace Analyses

• Collect traces
• where to monitor, how to collect (e.g.,
  efficiency, privacy, accuracy)
• Determine key metrics to characterize
• Process traces
• Draw inferences from the data
• Apply the traces or insights gained from the
  trace analyses to design better protocols
  systems

16
Common Analysis Techniques - Statistics

• Mean
• Median
• Geometric mean less sensitive to outliers
• Variance and standard deviation
• Confidence interval
• A range of values that has a specified
  probability of containing the parameter being
  estimated
• Example 95 confidence interval 10 ? x ? 20

17
Common Analysis Techniques Statistics (Cont.)

• Cumulative distribution (CDF) (a, P(X ? a))
• Complementary CDF (a, P(xgta))
• Probability density function (PDF)
• Derivative of CDF f(x) dF(x)/dx
• Check for heavy tail distribution
• Log-log complementary plot, and check its tail
• Example Pareto distribution If ??2,
  distribution has infinite variance (a heavy
  tail)If ??1, distribution has infinite mean

18
Common Analysis Techniques Data Fitting

• Visually compare two distributions
• Chi Squared tests AS86,Jain91
• Divide the data points into k bins
• Compute
• If X2 ? X2(?,k-c), then two distributions are
  close, where ? is significance level, c is the
  number of estimated parameters for the
  distribution 1
• Need enough samples
• Kolmogorov-Smirnov tests AS86,Jain91
• Compares two distributions by finding the maximum
  differences between two variables cumulative
  distribution functions
• Need to fully specify the distribution

19
Common Analysis Techniques Data Fitting (Cont.)

• Anderson-Darling Test Ste74
• Modification of the Kolmogorov-Smirnov test,
  giving more weight to the tails
• If A ? critical value, two distributions are
  similar otherwise they are not (F is CDF, and Yi
  are ordered data)
• Quantile-quantile plots AS86,Jain91
• Compare two distributions by plotting the inverse
  of the cumulative distribution function F-1(x)
  for two variables, and find best fitting line
• If the slope of the line is close to 1, and
  y-intercept is close to 0, the two data sets are
  almost identically distributed

20
Common Analysis Tools

• Scripting languages
• VB, Perl, awk, UNIX shell scripts,
• Databases
• SQL, DB2,
• Statistics packages
• Matlab, S, R, SAS,
• Write our own low level programs
• C, C, C,

21
Challenges in Workload Characterization

• Workload characteristics vary both in space and
  in time
• Each of the Web components provides a limited
  perspective on the functioning of the Web

Internet
proxy
replica
proxy
replica
proxy
Servers
Clients
22
Workload Variation

• Vary with measurement points
• Vary with sites being measured
• Information servers (news site), e-commercial
  servers, query servers, streaming servers,
  upload servers
• US vs. Europe,
• Vary with the clients being measured
• Internet clients vs. wireless clients
• University clients vs. home users
• US vs. Europe,
• Vary in time
• Day vs. night
• Weekday vs. weekend
• Changes with new applications, recent events
• Evolve over time,

23
Different Web Components Views

• View from clients
• Know details of client activities, such as
  requests satisfied by browser caches, client
  aborts
• The ability to record detailed information
• View from servers
• Most requests to the servers, excluding those
  satisfied by browser proxy caches
• May not log detailed information to ensure fast
  processing of client requests
• View from proxies
• Depending on the proxys location
• A proxy close to clients see requests from a a
  small number of clients to a large number of
  servers KR00
• A proxy close to the servers see requests from a
  large number of clients to a small number of
  servers KR00
• Requests satisfied by browser caches or proxy
  caches encountered earlier will not appear in the
  logs

24
Part II Web Workload

• Overview
• Content dynamics
• Access dynamics
• Common pitfalls
• Case studies

25
Content Dynamics

• File types
• File size distribution
• File update patterns
• How often files are updated
• How much files are updated

26
File Types

• Text files
• HTML, plain text,
• Images
• Jpeg, gif, bitmap,
• Applications
• Javascript, cgi, asp, pdf, ps, gzip, ppt,
• Multimedia files
• Audio, video

27
File Size Distribution

• Two definitions
• D1 Size of all files on a Web server
• D2 Size of all files transferred by a Web server
• D1 ? D2, because some files can be transferred
  multiple times or not in completion and other
  files are not transferred
• Studies show that the distribution of file sizes
  in both definitions exhibit heavy tails (i.e.,
  PF gt x x-?, 0 ? ? ? 2)

28
File Update Interval

• Varies in time
• Hot events and fast changing events require more
  frequent update, e.g., Worldcup
• Varies across sites
• Depending on server update policies update
  tools
• Depending on the nature of content (e.g.,
  University sites have slower update rate than
  news sites)
• Recent studies
• Study of the proxy traces collected at DEC and
  ATT in 1996 showed the rate of file change
  depended on content type, top-level domains etc.
  DFK97
• Study of 1999 MSNBC logs shows that modification
  history yields a rough predictor of future
  modification interval PQ00

29
Extent of Change upon Modifications

• Varies in time
• Different events trigger different amount of
  updates
• Varies across sites
• Depending on servers update policies and update
  tools
• Depending on the nature of the content
• Recent studies
• Studies of 1996 DEC and ATT proxy MDF97 and
  1999 MSNBC log PQ00 show that most file
  modifications are small ? delta encoding can be
  very useful

30
Part II Web Workload

• Motivation
• Limitations of workload measurements
• Content dynamics
• Access Dynamics
• Common pitfalls
• Case studies

31
Access Dynamics

• File popularity distribution
• Temporal stability
• Spatial locality
• User request arrivals durations

32
Document Popularity

• Web requests follow Zipf-like distribution
• Request frequency ? 1/i?, where i is a
  documents ranking
• The value of ? depends on the point of
  measurements
• Between 0.6 and 1 for client traces and proxy
  traces
• Close to or larger than 1 for server traces
  ABC96, PQ00
• The value of ? varies over time (e.g., larger ?
  during hot events)

33
Impact of the value ?

• Larger ? means more concentrated accesses on
  popular documents ?caching is more beneficial
• 90 of the accesses are accounted by
• Top 36 files in proxy traces BCF99, PQ00
• Top 10 files in small departmental server logs
  reported in AW96
• Top 2-4 files in MSNBC traces

34
Temporal Stability

• Metrics
• Coarse-grained likely duration that a set of
  current popular files remain popular
• e.g., overlap between the set of popular
  documents on day 1 and day 2
• Fine-grained how soon a requested file will be
  requested again
• e.g., LRU stack distance ABC96

File 5
File 2
File 4
File 5
Stack distance 4
File 3
File 4
File 2
File 3
File 1
File 1
35
Spatial Locality

• Refers to if users in the same geographical
  location or at the same organization tend to
  request a similar set of content
• E.g., compare the degree of requests locally
  shared

36
Spatial Locality (Cont.)
Domain membership is significant except when
there is a hot event of global interest
37
User Request Arrivals Duration

• User workload at three levels
• Session a consecutive series of requests from a
  user to a Web site
• Click a user action to request a page, submit a
  form, etc.
• Request each click generates one or more HTTP
  requests
• Exponential distribution LNJV99,KR01
• Session duration
• Heavy-tail distribution KR01
• clicks in a session, most in the range of 4-6
  Mah97
• embedded references in a Web page
• Think time time between clicks
• Active time time to download a Web page and its
  embedded images

38
Common Pitfalls

• Trace analyses are all about writing scripts
  plotting nice graphs
• Challenges
• Trace collection where to monitor, how to
  collect (e.g., efficiency, privacy, accuracy)
• Identify important metrics, and understand why
  they are important
• Sound measurements require disciplines Pax97
• Dealing with errors and outliers
• Draw implications from data analyses
• Understanding the limitation of the traces
• No representative traces workload changes in
  time and in space
• Try to diversify data sets (e.g., collect traces
  at different places and different sites) before
  jumping into conclusions
• Draw inferences more than what data show

39
Part II Web Workload

• Motivation
• Limitations of workload measurements
• Content dynamics
• Access dynamics
• Common pitfalls
• Case studies
• Boston University client log study
• UW proxy log study
• MSNBC server log study
• Popular Mobile server log study

40
Case Study I BU Client Log Study

• Overview
• One of the few client log studies
• Analyze clients browsing pattern and their
  impact on network traffic CBC95
• Approaches
• Trace collection
• Modify Mosaic and distribute it to machines in CS
  Dept. at Boston Univ. to collect client traces in
  1995
• Log format ltclient machine, request time, user
  id, URI, document size, retrieval timegt
• Data analyses
• Distribution of document size, document
  popularity
• Relationship between retrieval latency and
  response size
• Implications on caching strategies

41
Major Findings

• Power law distributions
• Distribution of document sizes
• Distribution of user requests for documents
• requests to documents as a function of their
  popularity
• Caching strategies should take into account of
  document size (i.e., give preference to smaller
  documents)

42
Case Study II UW Proxy Log Study

• Overview
• Proxy traces collected at the University of
  Washington
• Approaches WVS99a, WVS99b
• Trace collection deploy a passive network
  sniffer between the Univ. of Washington and the
  rest of the Internet in May 1999
• Set well-defined objectives
• Understand the extent of document sharing within
  an organization and across different
  organizations
• Understand the performance benefit of cooperative
  proxy caching

43
Major Findings

• Members of an organization are more likely to
  request the same documents than a random set of
  clients
• Most popular documents are globally popular
• Cooperative caching is most beneficial for small
  organizations
• Cooperative caching among large organizations
  yield minor improvement if any

44
Case Study III MSNBC Server Log Study

• Overview of MSNBC server site
• a large news site
• server cluster with 40 nodes
• 25 million accesses a day (HTML content alone)
• Period studied Aug. Oct. 99 Dec. 17, 98
  flash crowd

45
Approaches

• Trace collection
• HTTP access logs
• Content Replication System (CRS) logs
• HTML content logs
• Data analyses
• Content dynamics
• How often files are modified?
• How to predict modification interval?
• How much does a file change upon modification?
• Access dynamics
• Document popularity
• Temporal stability
• Spatial locality
• Correlation between document age and popularity

46
Major Findings

• Content dynamics
• Modification history is a rough predictor ? guide
  for setting TTL, but need an alternative
  mechanism (e.g., callback based invalidation) as
  backup
• Frequent but minimal file modifications ? delta
  encoding
• Access dynamics
• Set of popular files remains stable for days ?
  pushing/prefetching previous hot data that have
  undergone modifications
• Domain membership has a significant bearing on
  client accesses except during a flash crowd of
  global interest ? make sense to have a proxy
  cache for an organization
• Zipf-like distribution of file popularity but
  with a much larger ? than at proxies ? potential
  of reverse caching and replication

47
Case Study IV Popular Mobile Server Log

• Overview of a popular commercial Web site for
  mobile clients
• Content
• news, weather, stock quotes, email, yellow pages,
  travel reservations, entertainment etc.
• Services
• Notification
• Browse
• Period studied
• 3.25 million notifications in Aug. 20 26, 2000
• 33 million browse requests in Aug. 15 26, 2000

48
Approaches

• Analyze by user categories
• Cellular users
• Browse the Web in real time using cellular
  technologies
• Offline users
• Download content onto their PDAs for later
  (offline) browsing, e.g. AvantGo
• Desktop users
• Signup services and specify preferences
• Analyze by Web services
• Browse
• Notifications
• Use SQL database to manage data

49
Major Findings

• Notification Services
• Popularity of notification messages follows a
  Zipf-like distribution, with top 1 notification
  objects responsible for 54-64 of total messages
  ? multicast notifications
• Exhibits geographical locality ? useful to
  provide localized notification services
• Browse Services
• 0.1 - 0.5 queries account for 90 requests ?
  cache the results of popular queries
• The set of popular queries remain stable ? cache
  a stable set of queries or optimize query based
  on a stable workload
• Correlation between the two services
• Correlation is limited ? influence design of
  pricing plans

50
Tutorial Outline

• Background
• Web Workload
• Performance Diagnosis
• Applications of traces

51
Part III Performance Diagnosis

• Overview of performance diagnosis
• Infer the causes of high end-to-end delay in Web
  transfers BC00
• Infer the causes of high end-to-end loss rate in
  Web transfers CDH99,DPP01,NC01,PQ02, PQW02

52
Overview of Performance Diagnosis

• Goal determine trouble spot locations
• Metrics of interest
• Delay
• Loss rate
• Raw bandwidth
• Available bandwidth
• Traffic rate
• Why interesting
• Resolve the trouble spots
• Server selection
• Placement of mirror servers

Web Server
ATT
Sprint
MCI
UUNET
Earthlink
Why so slow?
AOL
Qwest
53
Finding the Sources of Delays

• Goal
• Why is my Web transfer slow? Is it because of the
  server or the network or the client?
• Sources of delay in Web transfer
• DNS lookup
• Server delays
• Client delays
• Network delays
• Propagation delays
• Queuing delays
• Delays introduced by packet losses (e.g.,
  signaled by the fast retransmit mechanism or TCP
  timeouts)

54
TCPEval Tool

• Inputs tcpdump packet traces taken at the
  communicating Web server and client
• Generates a variety of statistics for file
  transactions
• File and packet transfer latencies
• Packet drop characteristics
• Packet and byte counts per unit time
• Generates both timeline and sequence plots for
  transactions
• Generates critical path profiles and statistics
  for transactions

55
Critical Path Analysis Tool BC00
Data flow
Critical Path
Server
Client
Server
Client
Network delay
Network delay
Server delay
Network delay
Client delay
Network delay
Server delay
Network delay due to pkt loss
56
Finding Sources of Packet Losses

• Goal
• Identify lossy links

• (1-l1)(1-l2)(1-l4) (1-p1)
• (1-l1)(1-l2)(1-l5) (1-p2)
• (1-l1)(1-l3)(1-l8) (1-p5)
• Challenges
• an under-constrained system of equations
• - measurement errors

server
l1
l3
l2
l8
l7
l6
l4
l5
clients
p1
p2
p3
p4
p5
57
Approaches

• Active probing
• Probing
• Multicast probes
• Striped unicast probes
• Technique -- Expectation Maximization (EM)
  CDH99, DPP01
• a numerical algorithm to compute ? that maximizes
  P(D?), where D are observations, ? are ensemble
  of link loss rates

58
Approaches (Cont.)

• Passive monitoring
• Random sampling
• Random sample the solution space, and draw
  conclusions based on samples
• Akin to Monte Carlo sampling
• Linear optimization
• Determine a unique solution by optimizing an
  objective function
• Gibbs sampling
• Determine P(?D) by drawing samplings, where ? is
  ensemble of loss rates of links in the network,
  and D is observed packet transmission and losses
  at the clients
• EM
• A numerical algorithm to compute ? that maximizes
  P(D?)

59
Other Performance Studies using Web traces

• Characterize Internet performance (e.g., spatial
  temporal locality) BSS97
• Study the behavior of TCP during Web transfers
  BPS98

60
Tutorial Outline

• Background
• Web Workload
• Performance Diagnosis
• Applications of traces
• Bibliography

61
Part IV Applications of Traces

• Synthetic workload generation
• Cache design
• Cache replacement policies CI97,BCF99
• Cache consistency algorithms LC97, YBS99,YAD01
• Cooperative cache or not WVS99
• Cache infrastructure
• Pre-fetching algorithms CB98, FJC99
• Placement of Web proxies/replicas QPV01
• Other optimizations
• Improving TCP for Web transfers
  Mah97,PK98,ZQK00
• Concurrent downloads, pipelining, compression,

62
Synthetic Workload Generation

• Generate user requests
• Generate user sessions using a Poisson arrival
  process
• For each user session, determine clicks using a
  Pareto distribution
• Assign a click to a request for a Web page, while
  making sure
• The popularity distribution of files follows a
  Zipf-like distribution BC98
• Capture the temporal locality of successive
  requests for the same resource
• Generate a next click from the same user with
  think time following a Pareto distribution

63
Synthetic Workload Generation (Cont.)

• Generate Web pages
• Determine the number of Web pages
• Generate the size of each Web pages using a
  log-normal distribution
• Associate a page with some number of embedded
  pages using an empirical distribution
  (heavy-tail)
• Generate file modification events
• Examples of generators
• Webbench Wbe, WebStoneTS95, Surge BC98,
  SPecweb99 SP99, Web Polygraph WP,

64
Cache Replacement Policies

• Problem formulation
• Given a fixed size cache, how to evict pages to
  maximize the hit ratio once the cache is full?
• Hit ratio
• Fraction of requests satisfied by the cache
• Fraction of the total size of requested data
  satisfied by the cache
• Factors to consider
• Request frequency
• Modification frequency
• Benefit of caching reduction in latency BW
• Cost of caching storage
• Caveat NOT all hits are equal. Hit ratios do NOT
  map directly to performance improvement.

65
Cache Replacement Policies (Cont.)

• Approaches
• Least recently used (LRU)
• Least frequently used (LFU)
• Perfect maintain counters for all pages seen
• In-cache maintain counters only for pages that
  are in cache
• GreedyDual-size CI97
• Assign a utility value to each object, and
  replace the one with the lowest utility
• Use of traces
• Evaluate the algorithms using trace-driven
  simulations
• Analytically derive the hit ratios for different
  replacement policies based on a workload model

66
References

• AS86 R. B. DAgostino and M. A. Stephens.
  Goodness-of-Fit Techniques. Marcel Dekker, New
  York, NY 1986.
• ABC96 Virgilio Almeida, Azer Bestavros, Mark
  Crovella and Adriana de Oliveria. Characterizing
  reference locality in the WWW. In Proceedings of
  1996 International Conference on Parallel and
  Distributed Information Systems (PDIS'96),
  December 1996.
• ABQ01 A. Adya, P. Bahl, and L. Qiu. Analyzing
  Browse Patterns of Mobile Clients. In Proc. of
  SIGCOMM Measurement Workshop, Nov. 2001.
• ABQ02 A. Adya, P. Bahl, and L. Qiu.
  Characterizing Alert and Browse Services for
  Mobile Clients. In Proc. of USENIX, Jun. 2002.
• AL01 P. Albitz, and C. Liu. DNS and BIND (4th
  Edition), OReilly Associates, Apr. 2001.
• AW97 M. Arlitt and C. Williamson. Internet Web
  Servers Workload Characterization and
  Performance Implications. IEEE/ACM Transactions
  on Networking, Vol. 5, No. 5, pp. 631-645,
  October 1997.
• BC98 P. Barford and M. Crovella. Generating
  representative workloads for network and server
  performance evaluation. In Proc. of SIGMETRICS,
  1998.

67
References (Cont.)

• BBC98 P. Barford, A. Bestavros, M. Crovella,
  and A. Bradley. Changes in Web Client Access
  Patterns Characteristics and Caching
  Implications, Special Issue on World Wide Web
  Characterization and Performance Evaluation
  World Wide Web Journal, December 1998.
• BCF99 L. Breslau, P. Cao, L. Fan, G. Phillips,
  and S. Shenker. Web Caching and Zipf-like
  Distributions Evidence and Implications. In
  Proc. of INFOCOM, Mar. 1999.
• BC00 P. Barford and M. Crovella. Critical Path
  Analysis of TCP Transactions. In Proc. of ACM
  SIGCOMM, Aug. 2000.
• BLFF96 T. Berners-Lee, R. Fielding, and H.
  Frystyk. Hypertext Transfer Protocol -- HTTP/1.0.
  RFC 1945, May 1996.
• BPS98 H. Balakrishnan, V. N. Padmanabhan, S.
  Seshan, M. Stemm and R. H. Katz. TCP Behavior of
  a Busy Internet Server Analysis and
  Improvements. In Proc. IEEE Infocom, San
  Francisco, CA, USA, March 1998.
• BSS97 H. Balakrishnan, S. Seshan, M. Stemm,
  and R. H. Katz. Analyzing Stability in Wide-Area
  Network Performance. In Proc. of SIGMETRICS, Jun.
  1997.

68
References (Cont.)

• CDH99 R. Caceres, N. G. Duffield, J. Horowitz,
  D. Towsley, T. Bu. Multicast-Based Inference of
  Network Internal Loss Characteristics. In Proc.
  Infocom, Mar. 1999.
• CB98 M. Crovella and P. Barford. The network
  effects of prefetching. In Proc. of INFOCOM,
  1998.
• CBC95 C. R. Cunha, A. Bestavros, and M. E.
  Crovella. Characteristics of WWW client-based
  traces. Technical Report BU-CS-95-010, CS Dept.,
  Boston University, 1995.
• CI97 P. Cao and S. Irani. Cost-Aware WWW proxy
  caching algorithms. In Proc. of USITS, Dec. 1997.
• DFK97 F. Douglis, A. Feldmann, B.
  Krishnamurth, and J. Mogul. Rate of change and
  other metrics a live study of the World Wide
  Web. In Proc. of USITS, 1997.
• DPP01 N. G. Duffield, F. Lo Presti, V. Paxson,
  D. Towsley. In Proc. Infocom, Apr. 2001.
• FCD99 A. Feldmann, R. Caceres, F. Douglis, and
  M. Rabinovich. Performance of Web Proxy Caching
  in heterogeneous bandwidth enviornments. In Proc.
  of INFOCOM, March 1999.

69
References (Cont.)

• FJC99 L. Fan, Q. Jacobson, P. Cao and W. Lin.
  Web Prefetching Between Low-Bandwidth Clients and
  Proxies Potential and Performance. In Proc. of
  SIGMETRICS, 1999.
• FCT02 Y. Fu, L. Cherkassova, W. Tang, and A.
  Vahdat. EtE Passive End-to-End Internet Service
  Performance Monitering. In Proc. of USENIX, Jun.
  2002.
• GMF99 J. Gettys, J. Mogul, H. Frystyk, L.
  Masinter, P. Leach, T. Berners-Lee. Hypertext
  Transfer Protocol HTTP 1.1. RFC 2616, Jun.
  1999.
• JK88 V. Jacobson, M. J. Karels. Congestion
  Avoidance and Control. In Proc. SIGCOMM, Aug.
  1988.
• JJK01 S. Jamin, C. Jin, A. R. Kurc, D. Raz,
  and Y. Shavitt. Constrained Mirror Placement on
  the Internet. In Proc. of INFOCOM, Apr. 2001.
• Jain91 R. Jain. The Art of Computer Systems
  Performance Analysis. John Wiley and Sons, 1991.
• Kel02 T. Kelly. Thin-Client Web Access
  Patterns Measurements from a Cache-Busting
  Proxy. Computer Communications, Vol. 25, No. 4
  (March 2002), pages 357-366. 
• KR01 B. Krishnamurthy and J. Rexford. Web
  Protocols and Practice, HTTP/1.1, Networking
  Protocols, Caching, and Traffic Measurement.
  Addison-Wesley, May 2001.

70
References (Cont.)

• LC97 C. Liu and P. Cao. Maintaining Strong
  Cache Consistency in the World-Wide Web. In Proc.
  of ICDCS'97, pp. 12-21, May 1997.
• LNJV99 Z. Liu, N. Niclausse, and C.
  Jalpa-Villaneuva. Web Traffic Modeling and
  Performance Comparison Between HTTP 1.0 and HTTP
  1.1. In Erol Gelenbe, editor, System Performance
  Evaluation Methodologies and Applications. CRC
  Press, Aug. 1999.
• Mah97 Bruce Mah. An empirical model of HTTP
  network traffic. In Proc. of INFOCOM, April 1997.
• Mogul95 Jeffrey C. Mogul. The Case for
  Persistent-Connection HTTP. In Proc. SIGCOMM '95,
  pages 299-313. Cambridge, MA, August, 1995.
• MDF97 J. C. Mogul, F. Douglis, A. Feldmann,
  and B. Krishnamurthy. Potential benefits of
  delta-encoding and data compression for HTTP, In
  Proc. of SIGCOMM, September 1997.
• NC01 R. Nowak and M. Coates. Unicast Network
  Tomography using the EM algorithm. Submitted to
  IEEE Transactions on Information Theory, Dec.
  2001
• Pad95 V. N. Padmanabhan. Improving World Wide
  Web Latency. Technical Report UCB/CSD-95-875,
  University of California, Berkeley, May 1995.

71
References (Cont.)

• PQ00 V. N. Padmanabhan and L. Qiu. The Content
  and Access Dynamics of a Busy Web Server. In
  Proc. of SIGCOMM, Aug. 2000.
• PQ02 V. N. Padmanabhan and L. Qiu. Network
  Tomography using Passive End-to-End Measurements,
  DIMACS on Internet and WWW Measurement, Mapping
  and Modeling, Feb. 2002.
• PQW02 V. N. Padmanabhan, L. Qiu, and H. J.
  Wang. Passive Network Tomography using Bayesian
  Inference. Internet Measurement Workshop, Nov.
  2002.
• QPV01 L. Qiu, V. N. Padmanabhan, and G. M.
  Voelker. On the Placement of Web Server Replicas.
  In Proc. of INFOCOM, Apr. 2001.
• SP99 SPECWeb99 Benchmark. http//www.spec.org/os
  g/web99/.
• Pax98 V. Paxson. An Introduction to Internet
  Measurement and Modeling. SIGCOMM98 tutorial,
  August 1998.
• Ste74 M. A. Stephens. EDF Statistics for
  Goodness of Fit and Some Comparison. Journal of
  the American Statistical Association, Vol. 69,
  pp. 730 737.
• TS95 G. Trent and M. Sake. WebStone The First
  Generation in HTTP Server Benchmarking, Feb.
  1995. http//www.mindcraft.com/webstone/paper.html
  .

72
References (Cont.)

• Wbe Webbench. http//www.zdnet.com/etestinglabs/
  stories/benchmarks/0,8829,2326243,00.html.
• WP Web Polygraph Proxy performance benchmark.
  http//polygraph.ircache.net/.
• WVS99a A. Wolman, G. Voelker, N. Sharma, N.
  Cardwell, M. Brown, T. Landray,D. Pinnel, A.
  Karlin, and H. Levy. Organization-Based Analysis
  of Web-Object Sharing and Caching. In Proc. of
  the Second USENIX Symposium on Internet
  Technologies and Systems, Boulder, CO, October
  1999.
• WVS99b A. Wolman, G. M. Voelker, N. Sharma, N.
  Cardwell, A. Karlin, and H. M. Levy. On the scale
  and performance of cooperative Web proxy caching.
  In Proc. of the 17th ACM Symposium on Operating
  Systems Principles, Kiawah Island, SC, Dec. 1999.
• YAD01 J. Yin, L. Alvisi, M. Dahlin, A. Iyengar.
  Engineering server-driven consistency for large
  scale dynamic services.
• YBS99 H. Yu, L. Breslau, and S. Shenker. A
  Scalable Web Cache Consistency Architecture. In
  Proc. of SIGCOMM, August 1999.

73
Acknowledgement

• Thank Alec Wolman for his helpful
• comments.

74
Thank you!\\valleyview\pub\internal-web-perf.ppt