1. Reducing Network Latency Paul Johnson CPSC 550 21 APRIL 05 Using an Intelligent Service to Determine the Cheapest Communications Path

2. Problem The von Neumann bottleneck is made worse by networking The von Neumann bottleneck is made worse by networking The networked information is just a new level in the memory hierarchy The networked information is just a new level in the memory hierarchy There is evidence that the latency for communication is not consistent, even when looking for the same server There is evidence that the latency for communication is not consistent, even when looking for the same server

3. Objective Create an intelligent service capable of reducing the overall communication latency by providing the best known routing path to the operating system on demand. Create an intelligent service capable of reducing the overall communication latency by providing the best known routing path to the operating system on demand.

4. Connectionless Communication Observing out of order datagram reception over a local network Observing out of order datagram reception over a local network

5. Monitoring Data Transfers (1) PING returns the latency when communicating with another computer PING returns the latency when communicating with another computer

6. Monitoring Data Transfers (2) PING can help recognize long response times PING can help recognize long response times

7. Monitoring Data Transfers (3) TRACERT displays the hops from point to point on the internet TRACERT displays the hops from point to point on the internet

8. Monitoring Data Transfers (4) TRACERT again goes to a new destination, and takes different hops TRACERT again goes to a new destination, and takes different hops

9. The Windows Way ROUTE PRINT shows the METRIC used to determine which route to use sending data ROUTE PRINT shows the METRIC used to determine which route to use sending data

10. Proposed Solution (1) Addition of a service to monitor possible routes to common servers Addition of a service to monitor possible routes to common servers The service would provide information to the OS on demand regarding the best path to a destination The service would provide information to the OS on demand regarding the best path to a destination Nodes at hop destinations could re-route the packet based on more complete information Nodes at hop destinations could re-route the packet based on more complete information

11. Proposed Solution (2) The service would know the best route from User to Server The service would know the best route from User to Server

12. Possible Implementation Runs in the background, taking advantage of unused clock cycles Runs in the background, taking advantage of unused clock cycles AI Expert System using rules based on consistency, time of return, and number of hops required AI Expert System using rules based on consistency, time of return, and number of hops required Testing packets created by the service can be dropped at any given time if “real work” becomes available Testing packets created by the service can be dropped at any given time if “real work” becomes available Partial path to the destination is maintained within the message Partial path to the destination is maintained within the message Entirely theoretical design, may require new types of hardware to support the service Entirely theoretical design, may require new types of hardware to support the service

13. Final Remarks Based on time of day, server load, and priority, each node may begin to have favorite paths – reducing contention for resources Based on time of day, server load, and priority, each node may begin to have favorite paths – reducing contention for resources This is a dynamic solution, capable of handling changes to networks This is a dynamic solution, capable of handling changes to networks The solution handles hardware failures without additional intervention The solution handles hardware failures without additional intervention

14. References Coulouris, George; Dollimore, Jean; and Kindberg, Tim. Distributed Systems: Concepts and Design, 3rd Edition. New York: Addison-Wesley, 2001. Coulouris, George; Dollimore, Jean; and Kindberg, Tim. Distributed Systems: Concepts and Design, 3rd Edition. New York: Addison-Wesley, 2001. Hennessy, John L. and Patterson, David A. Computer Architecture: A Quantitative Approach, 3rd Edition. San Francisco, California: Morgan Kaufmann Publishers, 2003. Hennessy, John L. and Patterson, David A. Computer Architecture: A Quantitative Approach, 3rd Edition. San Francisco, California: Morgan Kaufmann Publishers, 2003. Giarratano, Joseph. Expert Systems: Principles and Programming, 3rd Edition. Boston, Maryland: PWS Publishing Company, 1998 Giarratano, Joseph. Expert Systems: Principles and Programming, 3rd Edition. Boston, Maryland: PWS Publishing Company, 1998 Jackson, Peter. Introduction to Expert Systems, 3rd Edition. New York: Addison-Wesley, 1999. Jackson, Peter. Introduction to Expert Systems, 3rd Edition. New York: Addison-Wesley, 1999. Keogh, Jim. The Essential Guide to Networking. Upper Saddle River, New Jersey: Prentice-Hall, Inc., 2001. Keogh, Jim. The Essential Guide to Networking. Upper Saddle River, New Jersey: Prentice-Hall, Inc., 2001. Miller, Mark. Internet Technologies Handbook: Optimizing the IP Network. Hoboken, New Jersey: John Wiley & Sons, 2004. Miller, Mark. Internet Technologies Handbook: Optimizing the IP Network. Hoboken, New Jersey: John Wiley & Sons, 2004.