FREE Registration is required
Overview:
This paper first studies the accuracy of two well-known analytical models of the average throughput of long-term TCP flows, namely the so-called SQRT and PFTK models, and show that these models are far from being accurate in general. The simulations, based on a large set of long-term TCP sessions, show that 70% of their predictions exceed the boundaries of TCP-Friendliness, thus questioning their use in the design of new TCP-Friendly transport protocols. The paper then investigates the reasons of this inaccuracy, and show that it is largely due to the lack of discrimination between the two packet loss detection methods used by TCP, namely by triple duplicate acknowledgments or by timeout expirations.
(Is this item miscategorized? Does it need more tags? Let us know.)
| Format: | Size: | 743 KB | |
| Date: | Dec 2006 | ||
| Pages: | 25 |
White Papers, Webcasts, and Resources
- SIP Trunking is Key to Accelerating Unified Communications Deployments SprintConsidering unified communications (UC) for your business? Discover the key to simplifying deployment, maximizing value and raising adoption company-wide.
- Windows Phones and Unified Communications MicrosoftGain a more solid understanding of UC, why its essential for your business today, and what makes Windows phones ideal for secure UC environments.
- Enterprise and Web 2.0 application support in a modern mainframe environment IBMSee how IBM WebSphere Portal software can help you develop a Web presence based on individual needs while unlocking value for customers and employees.









