J. Border, M. Kojo, J. Griner, G. Montenegro, and Z. Shelby "Performance enhancing proxies," Internet Draft: http://community.roxen.com/developers/idocs/drafts/draft-ietfpilc -pep-04.html (accessed June 2003).  Home/Search   Document Not in Database   Summary   Related Articles   Check

....static client. Scenario 2: A File Transfer Session with the Mobile Client. As the mobile client is accessing the network via a wireless link, to improve the performance of the TCP connection between the file server and the mobile client, performance enhancing proxies are required. According to [2], over a wireless link, performance enhancing proxies not only can improve the performance of the link, but also keeping the TCP connection alive while the mobile client is disconnected from the network for a period of time. In this scenario, Berkeley s Snoop protocol [2] can be used to improve ....

....required. According to [2] over a wireless link, performance enhancing proxies not only can improve the performance of the link, but also keeping the TCP connection alive while the mobile client is disconnected from the network for a period of time. In this scenario, Berkeley s Snoop protocol [2] can be used to improve the performance of the wireless link. A Snoop proxy is placed in the last hop router to the mobile client. From Figure 1, the last hop to the mobile client is Compute Server 2. During file transfer, the Snoop proxy caches any unacknowledged TCP data and forwards it to the ....

Border J, Kojo M, Griner J, Montenegro G, "Performance Enhancing Proxies", IETF Internet-Draft, ietf-pilcpep. 00.txt

.... viable interworking scheme may consist in splitting end to end TCP connections into different (both terrestrial and satellite) connection segments, in order to de couple high latency or lossy network segments, e.g. the satellite one, from the rest of the network in a way transparent to end users [7]. Given space limitations, we assume that the reader is familiar with TCP Sat issues and TCP splitting techniques and omit relevant background discussions (in this regard, 15] is a useful reference) III.ESW INTERWORKING PRM Taking into account what summarised in section II, we propose a ....

....5, the M ESW gateway proxy implements connection splitting at IWL TCP layer level, to shield end users from the effect of delay and bit errors on the satellite link. The proxy is a transparent active TCP Performance Enhancing Proxy (PEP) intended to be deployed at the ends of the satellite hop [7]. This way, the two proxies split the end to end TCP connection into three separate connection segments . For example, by considering data flowing from end host H1 to end host H2, the IWL TCP of the M ESW gateway (G1) fakes the IP address of the remote end host (H2) acknowledges the opposite ....

RFC 3135, J. Border, M. Kojo, J. Griner, G. Montenegro, Z. Shelby: "Performance Enhancing Proxies", June 2001.

....DiffServ classes. Performance Enhancing Proxies (PEP) A Performance Enhancing Proxy (PEP) may be used to improve the performance of the Internet protocols on network path where native performance suffer due to the characteristics of the path (long delay and low channel reliability in satellites) [4] itself. There exist many types of PEPs to be used in different environments. Transport layer PEPs that interact with TCP is called TCP PEP. These proxies may be classified in two categories depending on their relationship with the end user. User visible proxies require some explicit configuration ....

J. Border, et al, Performance Enhancing Proxies, IETF Internet Draft. Work in progress.

....available bandwidth by dropping selected frames, and distilling [13] in which the resolution or color palette of the image is reduced to reduce the transmission time. Yet another class of proxies are designed to improve various aspects of TCP. The Internet draft on Performance Enhancing Proxies [6] describes several such proxies, including relatively simple functions like TCP ACK Spacing [3] 23] this proxy eliminates bursts of TCP data by smoothing out the flow of TCP ACKs) and TCP Snoop [4] this proxy hides packet drops on lossy links from the TCP sender to avoid wrongly triggering ....

J. Border, M. Kojo, Jim Griner, G. Montenegro, and Z. Shelby. Performance Enhancing Proxies. Internet-Draft (draft-ietf-pilcpep -04), October 2000.

.... many modifications and new solutions have been proposed to improve TCP s performance, such as forward error correction schemes, retransmissions at the link layer, split connections like MTCP [11] Explicit Loss Notification [7] link layer TCPAware like Snoop [4] Performance Enhancing Proxies [5], Indirect TCP (I TCP) 2] MAITE [1] etc. Available performance evaluations of TCP over wireless networks are usually incomplete, meaning that the most important TCP versions and the most important solutions are not studied and compared all together. Therefore, we still don t have a good idea ....

J. Border, M. Kojo, J. Griner, G. Montenegro, and Z. Shelby. Performance Enhancing Proxies. IETF Internet draft (draftietf -pilc-pep-07.txt), June 2001.

....in the current Internet. For example, TCP over satellite links has revealed network utilization issues and TCP s undesirable bias against long RTT flows [4] Currently, these problems are mitigated using ad hoc mechanisms such as ack spacing, split connection [4] or performance enhancing proxies [8]. Simulation results similar to those in 5 support the above argument showing that, regardless of the queuing scheme, TCP s performance degrades significantly as either capacity or delay increases. This paper develops a novel protocol for congestion control that outperforms TCP in ....

J. Border, M. Kojo, J. Griner, and G. Montenegro. Performance enhancing proxies, Nov. 2000.

....conditions. We also note that this paper only presents investigations into the performance of TCP spoofnag. While the performance of such a mechanism is an important point when deciding whether or not to use a spooler, it is not the only consideration. We encourage network operators to read [RFC3135] to gain a better understanding of some of the non performance based considerations involved in deciding whether or not to use a spooler. The remainder of the paper includes Section 2, which discusses spoofing in greater depth, and Section 3, which outlines the actual simulation mechanics and ....

....areas for future work on this subject. 2 BACKGROUND In an attempt to mitigate the disadvantages of TCP over long latency links, researchers have been introducing performante enhancing proxies (PEPs) into networks. One such PEP that is currently being used in satellite networks is TCP spoofing [RFC3135, BB95, ASBD96]. The objective of spoof nag involves isolating the long latency link by introducing a middle agent which splits the TCP connection (see Figure 1) However, unlike a proxy cache, spoofing is transparent to both the sender and receiver. Thus, the middle agent, or spooler , takes on the personality ....

[Article contains additional citation context not shown here]

J. Border, M. Kojo, J. CJriner, G. Montenegro, and Z. Shelby, "Performance Enhancing Proxies", RFC 3135, June 2001

....the personality of both parties. The responsibility of the spoofer is to intercept, cache, and acknowledge data received by the sender and then forward that data to the receiver. As a result, spoofing does break the end to end semantic of TCP, however while this raises several philosophical issues [PILC ID], those issues are not the focus of this paper. Finally, it is worth noting that in our model data segments and connection teardowns are spoofed, while connection setup remains end toend. Figure 1: Satellite Spoofing 3 Simulation Overview 3.1 Terminology The following terms are explained below ....

J. Border, M. Kojo, J. Griner, G. Montenegro, and Z. Shelby, "Performance Enhancing Proxies", Internet-Draft, November 2000 draft-ietf-pilc-pep-05.txt

....rate of change of bandwidth. The buffers may not be able to handle this delay and or the bandwidth estimation may be stale. The result is that the TCP performance suffers. III. NEW ARCHITECTURAL SOLUTION Our approach for improving TCP performance is to deploy TCP Performance Enhance Proxies (PEP [4]) at the edges of TCP unfriendly networks (see Fig. 4) to control the TCP flows passing through them. The basic idea is for the PEP element at an edge router to monitor the available bandwidth of the network, and to manipulate the ACK packets of a passing thru TCP flow accordingly. By deliberately ....

J. Border et. al., "Performance enhancing proxies," RFC 3135, http://www.ietf.org/rfc/rfc3135.txt, July 2000.

....for wired networks. As a result, many of them do not work properly or efficiently on wireless networks where radio channels have different characteristics and devices vary greatly in size and capability. A popular approach for solving this problem is to make use of performance enhancing proxies [4], 16] In this case, mobile users communicate with the performance enhancing proxies which in turn exchange information with the destination servers on behalf of the mobile users. At the proxy, information from the original server is converted to make it suitable for the wireless environment, ....

....wireless networks. Some of these functions operate at the application level, so the proxy servers need to understand the context of specific applications; others operate at the transport layer or below and only need to deal with individual packets, without knowledge of how the applications work [4]. At the application level, Caching and prefetching are two basic mechanisms for reducing access latency [1] 8] Content transformation covers a broad range of operations that proxies may perform. The key idea is to convert data to a format that is more suitable to the mobile user, based on ....

[Article contains additional citation context not shown here]

J. Border, M. Kojo, J. Griner, and G. Montenegro, "Performance enhancing proxies," Internet-draft, http://search.ietf. org/internet-drafts/draft-ietf-pilc-pep-02.txt, March 2000.

No context found.

J. Border, M. Kojo, J. Griner, G. Montenegro, and Z. Shelby "Performance enhancing proxies," Internet Draft: http://community.roxen.com/developers/idocs/drafts/draft-ietfpilc -pep-04.html (accessed June 2003).

No context found.

J. Border, M. Kojo, J. Griner, G. Montenegro, "Performance Enhancing Proxies," Internet Engineering Task Force Draft, June 25, 1999 (see http://www.ietf.org/ under Internet Drafts).

No context found.

Border, J., Kojo, M., Griner, J., Montenegro, G. and Z. Shelby, "Performance Enhancing Proxies", RFC 3135, June 2001.

No context found.

J. Border, M. Kojo, J. Griner, G. Montenegro, and Z. Shelby, "Performance Enhancing Proxies", Internet Draft (work in progress), <draft-ietf-pilc-pep-04.txt>, October 2000.

No context found.

Border, J., Kojo, M., Griner, J., Montenegro, G. and Z. Shelby, "Performance Enhancing Proxies", RFC 3135, June 2001.

No context found.

M. Kojo, J. Griner, G. Montenegro, "Performance Enhancing Proxies", Internet Draft (draft-ietf-pilcpep -02.txt), Work in progress, March 2000

No context found.

J. Border, M. Kojo, J. Griner, G. Montenegro, and Z. Shelby. Performance enhancing proxies. RFC 3135, June 2001. 137 138

No context found.

J. Border, K. Kojo, J Griner, G. Montenegro, and Z. Shelby. Performance enhancing proxies. Internet draft, draft-ietf-pilc-pep-04.txt, work in progress, October 2000.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC