J. Crowcroft, I. Wakeman, Z. Wang, and D. Sirovica, "Is Layering Harmful?", IEEE Networks, Vol. 6, Issue 1, pp 20-24, January 1992. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Hardware-Assisted Networking Using Scheduled Transfer Protocol.. - Pietikäinen (2001) (Correct)
....transferring data is divided into multiple smaller problems. This causes problems in the implementations, since the individual layers have no knowledge about the overall picture and thus may make bad decisions about how to buffer data. This may cause severely degraded and unpredictable performance [6, 7, 8]. The alternative approach is practical only when the applications are running on the same machine or when using specialized cluster interconnect hardware, as then the requirement of an infinitely fast reliable network can be fulfilled. For almost any other kind of networking some abstraction ....
Crowcroft J., Wakeman I, Wang Z. & Sirovica D. (1992) Is Layering Harmful? IEEE Network Magazine, vol. 6, no. 1, pp. 20-24
Deadlock situations in TCP over ATM - Moldeklev, Gunningberg (1994) (11 citations) (Correct)
....the normal and degraded throughput is not that big, which makes it difficult to detect or uninteresting to investigate into. We know of one paper reporting on performance degradation due to the interaction between the delayed acknowledgment strategy and Nagle s algorithm, namely Crowcroft et al. [5]. They report on a boundary effect of remote procedure call (RPC) response time over TCP and ethernet. The most straightforward way to prevent many of the deadlock situations is to switch off Nagle s algorithm. As will be presented, there is hardly any performance penalty having it switched off. A ....
Crowcroft, J, Wakeman, I, Wang, Z, and Sirovica, D. Is layering harmful? IEEE Network, 6 (1), 20-24, January 1992.
A Component-Based Architecture For Software Communication Systems - Jung, Biersack (2000) (1 citation) (Correct)
....structure protocols. The advantage of layering is to reduce complexity of communication systems by hiding information between different layers. However, a number of problems with regard to layering have been identified such as inflexibility [35] inefficiency [33] and even unexpected side effects [12]. A lot of work tackles the efficiency problems of layering. Clark [7] and Cooper [10] propose to carefully break up the borders and allow adjacent layers to exchange control information. Clark [8] and Atkins [1] suggested to apply vertical process models to protocol software in order to reduce ....
J. Crowcroft, I. Wakeman, and Z. Wang. Is layering harmful ? IEEE Network, 6(1), Jan. 1992.
Deadlock situations in TCP over ATM - Moldeklev, Gunningberg (1994) (11 citations) (Correct)
....the normal and degraded throughput is not that big, which makes it difficult to detect or uninteresting to investigate into. We know of one paper reporting on performance degradation due to the interaction between the delayed acknowledgment strategy and Nagle s algorithm, namely Crowcroft et al. [7]. They report on a boundary effect of remote procedure call (RPC) response time over TCP and ethernet. The most straightforward way to prevent many of the deadlock situations is to switch off Nagle s algorithm. As will be presented, there is hardly any performance penalty having it switched off. A ....
Crowcroft, J, Wakeman, I, Wang, Z, and Sirovica, D. Is layering harmful? IEEE Network, 6 (1), 20-24, January 1992.
Distributed Network Computing over Local ATM Networks - Mengjou Lin (1994) (Correct)
....information for the message sizes around 8 Kbytes in greater detail, we also ran an experiment with message sizes that varied from 4 bytes to 16 Kbytes in 96 byte increments. The results are presented in Figure 8(b) This is a known TCP timing abnormality which has been reported by Crowcroft [9] and others. We have observed the same timing abnormality in our local ATM network. However, after enabling TCP NODELAY, this timing abnormality disappears. We performed experiments with the same TCP echo program, over both FDDI and Ethernet for message sizes less than 16 Kbytes. We found that ....
J. Crowcroft, I. Wakeman, Z. Wang, and D. Sirovica. Is Layering Harmful? IEEE Network, 6(1):20-- 24, January 1992.
Promela++: A Language for Constructing Correct and.. - Basu, Morrisett, von.. (Correct)
....Moreover, new protocols can be composed out of existing layers that have been previously tested and verified, thereby allowing protocol reuse. However, previous work has demonstrated that layering (and more generally, modularity) may add considerable overheads to protocol processing [CT90, CWWS92] Techniques for circumventing layering overheads have used various methods such as ILP [CT90] and microprotocols [OP92] that are not always easy to use or apply by hand. Therefore, the challenge is to develop an easily usable protocol development framework that combines the flexibility of ....
J. Crowcroft, I. Wakeman, Z. Wang, and D. Sirovica. Is Layering Harmful? IEEE Network, 6(1):20--24, January 1992.
Network Working Group R. Atkinson, Ed. Request for.. - Regarding Internet.. (Correct)
No context found.
J. Crowcroft, I. Wakeman, Z. Wang, and D. Sirovica, "Is Layering Harmful?", IEEE Networks, Vol. 6, Issue 1, pp 20-24, January 1992.
Unknown - Status Of This (Correct)
No context found.
J. Crowcroft, I. Wakeman, Z. Wang, and D. Sirovica, "Is Layering Harmful?", IEEE Networks, Vol. 6, Issue 1, pp 20-24, January 1992.
Unknown - (Correct)
No context found.
J. Crowcroft, I Wakeman, Z. Wang, & D. Sirovica, "Is Layering Harmful ?", IEEE Networks, January 1992.
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