Presentation is loading. Please wait.

Presentation is loading. Please wait.

Department of Computer Science Southern Illinois University Edwardsville Dr. Hiroshi Fujinoki and Kiran Gollamudi {hfujino,

Similar presentations


Presentation on theme: "Department of Computer Science Southern Illinois University Edwardsville Dr. Hiroshi Fujinoki and Kiran Gollamudi {hfujino,"— Presentation transcript:

1 Department of Computer Science Southern Illinois University Edwardsville Dr. Hiroshi Fujinoki and Kiran Gollamudi E-mail: {hfujino, kgollam}@siue.edu{hfujino,kgollam}@siue.edu

2 Response Time Problem Definition Problem The elapsed time between the end of an inquiry on a computer system and the beginning of a response Long response time Operating system overhead Due to high web traffic load FAT look-up, following a sector-chain, multiple clients etc Transmission time

3 Delay Causes We have 3 different causes of delay RECEIVER Network Protocol Processing Overhead Operating system overhead Internet Request Response Request (1) Response with tag (2) Request (3) Response (4) Internet Request Response SENDER Routing Overhead + Error/Flow control Overhead Internet

4 HTTP Client/Server model HTTP Client HTTP Server TCP Sync TCP Sync ACK HTTP Get Transmitting- requested file Requested file starts arriving. Time Response time Transmission time Terms Defined

5 Existing work for Client/Server model Existing Methods Caching Server Clustering Mirroring The following are the existing techniques to reduce response time and transmission delay

6 Object Packaging Objective Reduced network protocol overhead during the transmission at routers Reduced number of packets by minimizing fragmentation To improve response time and transmission time by : Single request for the multiple files Reduced OS overhead at a web server  Reduced FAT lookups  CRC calculation overhead  Memory copies  Flow control and error control

7 Concept Of Object Packaging Is a collection of web files in a web site Files are sequentially packed without compression Object Packaging: Object Package Number of objects Object Offset Fields File #1 File #n Pointer File header

8 Packaging Format Of Object Packaging Object Information Field File name sub File size subFile attribute sub Number of objects (files) sub Object Information Field File name subfield File size subfieldFile attribute subfield Number of objects (files) subfield Object Information Field: Contains the information of the packed files Data Field: Contains the contents of the requested files Number of objects: Tells how many packed files are there Sub field A collection of the names of the contained files A collection of the sizes of the contained files File Attribute Subfield: Each file is a binary, text, or executable

9 Experiment Modeling Client Machine: The host that requests the server for the data Hub : Broadcasts every packet to every port Traffic Monitor: Monitors the all the traffic (packets) that are passing Test-Bed Requests Requested Files Web Server Client Traffic Monitor Hub Local Disk Requests Requested Files Requests Requested Files Web Server Client Traffic Monitor Hub Local Disk

10 Experiment Design Files with 1K, 4K and 10K bytes are placed in the server machine Factors measured for both the existing method and object Packaging 2.Average bytes transferred 3. Number of Transferred packets 1. Average response time Experimental Setup 4. Transmission time

11 Experimental Output The figures represent all factors measured with different file sizes. Existing Method Object Packaging File 4K 0% 20% 40% 60% 80% 100% Avg. TimeNo.Of PktsBytes transferred Time Percentage File 10K Percentage Existing Method Object Packaging 0% 20% 40% 60% 80% 100% Avg. TimeNo.Of PktsBytes transferred

12 Experimental Output 0% 20% 40% 60% 80% 100% 1K4K10K Average file size (in bytes) Percentage to the existing method Transferred bytes Response time Transferred packets 0% 20% 40% 60% 80% 100% 1K4K10K Average file size (in bytes) Percentage to the existing method Transferred bytes Response time Transferred packets 0% 20% 40% 60% 80% 100% 0255075100 Percentage to the existing method Number of transferred files File size = 10K File size = 4K File size = 1K 0% 20% 40% 60% 80% 100% 0255075100 Percentage to the existing method Number of transferred files File size = 10K File size = 4K File size = 1K Number of transferred packets relative to the existing method

13 Summary No modification of an operating system at the server side nor transmission protocol at routers required Multiple file transmissions by object packaging proposed Object packaging is efficient in reducing response time and transmission load and time

14 Future work 1. Perform the experiments using a Gigabit Ethernet cable 2. To observe the scalability There are two on-going activities and future works 1. Measuring the CPU load, response time and propagation delay on the server 2. Measuring all the above but for multiple clients Future Works On Going Activities

15 References 1. M. Arlitt and C. Williamson, “Web Server Workload Characterization: The Search for Invariants,” Proceedings of the 1996 ACM SIGMETRICS Conference on the Measurement and Modeling of Computer Systems,” May 1996, pp. 126-137 2. GVU’s WWW User Surveys, Georgia Institute of Technology URL: http://www.gvu.gatech.edu/user_surveys 3. J. Ousterhout, “Why Aren't Operating Systems Getting Faster As Hardware?,” Proceedings of Summer 1990 USENIX Conference, June 1990, pp. 247-256 4. P. Druschel, “Operating System Support for High-Speed Networking,” Communications of the ACM, vol. 39, no. 2, September 1996, pp. 41-51 5. P. Markatos, “Speeding-up TCP/IP: Faster Processors Are not Enough,” Proceedings of the 21st IEEE International Performance, Computing, and Communications Conference, April 2002, pp. 341-345

16 References 6. M. Busari and C. Williamson, “On the Sensitivity of Web Proxy Cache Performance to Workload Characteristics,” Proceedings of IEEE INFOCOM,April 2001, pp. 1225-1234 7. J. Dilley, “The Effect of Consistency on Cache Response Time,” IEEE Network, vol. 14, no. 3, May/June 2000, pp. 24-28 8. S. Glassman, “A caching relay for the Worldwide Web,” Computer - Networks and ISDN Systems, vol. 27, no. 2, October 1994, pp. 165-173 9. D. Lee, “Pre­Fetch Document Caching to Improve Worldwide Web User Response Time,” Master's Thesis. Virginia Polytechnic Institute and State University, March 1996 10. J. Mogul, “Squeezing More Bits Out of HTTP Caches,” IEEE Network, vol. 14, no.3, May/June 2000, pp. 6-14 11. Figures of Red hat and Windows from the internet sites.

17 The End Thank you !!!

18 Client Side Caching Reduced latency Effective only when same files are repeatedly requested Reduced server load Reduced bandwidth consumption in a network Additional hardware or expertise is required No benefit if object is not cached May be unable to cache multimedia content Unable to cache dynamically generated content Advantages Disadvantages Go To Previous Slide

19 Mirroring Technique Improves throughput Low capacity Complete redundancy of data Fast recovery from a disk failure Expensive No improvement in data access speed Advantages Disadvantages Go To Previous Slide

20 Server Clustering Load balancing Fail over Fault resilience Scalability Requires investment for hardware Request dispatcher may be a bottleneck Disadvantages Advantages Go To Previous Slide


Download ppt "Department of Computer Science Southern Illinois University Edwardsville Dr. Hiroshi Fujinoki and Kiran Gollamudi {hfujino,"

Similar presentations


Ads by Google