D. Raz and Y. Shavitt. An Active Network Approach for Efficient Network Management. In IWAN'99, Berlin, May 1999.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....of router software in a real live network. However, given that the XORP router management system is fairly basic, we believe our system to be superior. There are a number of projects concerned with the management of programmable routers most notably NESTOR [37] DARWIN [20] SENCOMM [38] ABLE [39] and ANCORS [40] however there does not seem to be any management systems designed for networks consisting of multiple programmable routers. PONDER [41] a policy language that can be used to manage a wide variety of computing systems (including networks) is well suited for the management of ....

D. Raz and havitt, "An active network approach for efficient network management," in International Working Conference on Active Networks, Germany, 1999.

....my research extends work carried out in the domain of application level programmable networks. The key projects in this area are Promile [3] Android [18] and Click [19] There are few projects that are concerned with the management of programmable networks: NESTOR [4] SENCOMM [20] ABLE [21] and ANCORS [22] are among them. These projects seem to management systems for programmable routers rather than complete networks. One of the key components of the programmable network management middleware is the compiler that translates a declarative configuration policy into host specific ....

D. Raz, Y. Shavitt, "An Active network Approach for Efficient Network Management", in the proceedings of IWAN, Germany, 1999.

....[34] provides a platform for large scale Internet measurement using Java and Pythonbased modules. NIMI only allows active measurements, while the security policy and resource control issues are addressed using standard ACL like schemes. In the active networking arena, Smart Packets [39] and ABLE [36] provide programmable platforms for network management applications, following the MbD principles. ABLE allows applications to poll SNMP interfaces from inside the network, while Smart Packets operate on a management interface at the language level. In both cases, the management information and ....

Danny Raz and Yuval Shavitt. An active network approach for efficient network management. In IWAN'99, LNCS 1653, pages 220 --231, Berlin, Germany, 1999.

.... allowed per packet to some constant n, resulting in n O(jpj) O(jpj) The most conservative case would be to set n = 1, allowing only a single send or delivery per packet, matching unicast semantics but prohibiting multicast style programs, and even reasonable implementations of traceroute [6] [20]. A more flexible bound, used in our current implementation, is the least n such that multicast may be programmed: n varies per node to be the number of network interfaces on that node. For demux, we permit only one delivery per packet, and force the packet to exit following the delivery. As we ....

D. Raz and Y. Shavitt. An active network approach for efficient network management. In International Working Conference on Active Networks, July 1999.

....not be interpreted as representing the official policies, either expressed or implied, of the Defense Advanced Research Projects Agency or the U.S. Government. oped in 1994 and 1995 during discussions in the broad DARPA research community, and since then a significant number of prototypes (e.g. [10, 23, 24, 3, 18, 15]) have emerged. Reviewing the early AN discussions, we find one chief motivation driving the initiation of research into AN: faster network evolution. For example, the SwitchWare project proposal, from the University of Pennsylvania, states ( 21] p. 1) The pace of network evolution (not switch ....

....the elements that may be loaded are limited in the same way as Netscript boxes. 2.3.3 Other systems Some AN systems do not rely on dynamic linking as their underlying means of extensibility. Instead, they use a more traditional hardware based, process model for extensions. For example, ABLE [18] is an architecture whose extensions are processes spawned by the node s session manager. In these systems, plug ins are essentially whole programs whose extensibility interface consists of the allowed IPC mechanisms to the rest of the system. Just as plug ins are limited by the programming ....

Danny Raz and Yuvall Shavitt. An active network approach for efficient network management. In Stefan Covaci, editor, Proceedings of the First International Working Conference on Active Networks, volume 1653 of Lecture Notes in Computer Science, pages 220--231. Springer-Verlag, June 1999.

.... allowed per packet to some constant n, resulting in n O(jpj) O(jpj) The most conservative case would be to set n = 1, allowing only a single send or delivery per packet, matching unicast semantics but prohibiting multicast style programs, and even reasonable implementations of traceroute [6] [20]. A more flexible bound, used in our current implementation, is the least n such that multicast may be programmed: n varies per node to be the number of network interfaces on that node. For demux, we permit only one delivery per packet, and force the packet to exit following the delivery. As we ....

D. Raz and Y. Shavitt. An active network approach for efficient network management. In International Working Conference on Active Networks, July 1999.

.... many issues not discussed in this paper, including what language to use and how to distribute and compose active network programs [7] 32] 16] and how to make active networking secure [1] Applications of active networking include congestion control [2] caching [3] and network management [28]. IX. Conclusions Many network services can be evaluated satisfactorily only on a field trial. Unfortunately, today s networks still lack network programming platforms that would easily enable such trials. We argued that bridge configurations often enable convenient deployment, realistic ....

D. Raz and Y. Shavitt. "An Active Network Approach for Efficient Network Management," in Proc. IWAN'99, LNCS 1653, July 1999, pp. 220-231.

....triggers a computation in the receiving node, and communication is only allowed to immediate neighbors. The problem with the above model is that it fails to capture the way distributed algorithms are implemented in many modern network designs, e.g. high speed networks [4] and active networks [5]. Cidon et al. 6] were the first to address this problem in the context of high speed networks. They separated the delay associated with transmission and the delay associated with computation, and allowed messages to cutthrough a node with no computation delay penalty. However, their model is not ....

....different execution environments to perform tasks. It can also be used to evaluate the contribution of new features to an existing architecture. Our model captures an active IP network, i.e. the active nodes are part of a network that employs IPv4 routing. As in most currently suggested models [5], 7] packets arriving at a node (router) can be treated in two distinct ways: processed by some software centric process that is slow by nature, or forwarded via a fast track (called cutthrough in [7] that adds a negligible delay to the IP forwarding delay. The fast track, usually, involves ....

[Article contains additional citation context not shown here]

Danny Raz and Yuval Shavitt, "An active network approach for efficient network management," in IWAN'99, July 1999.

....in network elements as proposed in the active networks paradigm. Then, we use this framework to present and analyze distributed algorithms for basic problems that arise in the control and management of IP networks. Some of these algorithms were implemented in an active network test bed [14]. These problems include: route discovery, message dissemination, topology discovery, and bottleneck detection. In particular we show that for many needed tasks, adding computation power to the routers reduces the overall resource utilization. The suggested model can also be used to evaluate the ....

....triggers a computation in the receiving node, and communication is only allowed with immediate neighbors. The problem with the above model is that it fails to capture the way distributed algorithms are implemented in many modern network designs, e.g. high speed networks [3] and active networks [14]. Cidon et al. 4] were the first to address this problem in the context of high speed networks. They separated the delay associated with transmission and the delay associated with computation, and allowed messages to cut through a node with no computation delay penalty. However, their model is ....

[Article contains additional citation context not shown here]

D. Raz and Y. Shavitt. An active network approach for efficient network management. In IWAN'99, pages 220 -- 231, July 1999.

....in network elements as proposed in the active networks paradigm. Then, we use this framework to present and analyze distributed algorithms for basic problems that arise in the control and management of IP networks. Some of these algorithms were implemented in an active network test bed [14]. These problems include: route discovery, message dissemination, topology discovery, and bottleneck detection. In particular we show that for many needed tasks, adding computation power to the routers reduces the overall resource utilization. The suggested model can also be used to evaluate the ....

....triggers a computation in the receiving node, and communication is only allowed with immediate neighbors. The problem with the above model is that it fails to capture the way distributed algorithms are implemented in many modern network designs, e.g. high speed networks [3] and active networks [14]. Cidon et al. 4] were the first to address this problem in the context of high speed networks. They separated the delay associated with transmission and the delay associated with computation, and allowed messages to cut through a node with no computation delay penalty. However, their model is ....

[Article contains additional citation context not shown here]

D. Raz and Y. Shavitt. An active network approach for efficient network management. In IWAN'99, pages 220 -- 231, July 1999.

No context found.

D. Raz and Y. Shavitt. An Active Network Approach for Efficient Network Management. In IWAN'99, Berlin, May 1999.

No context found.

D. Raz and Y. Shavitt. An Active Network Approach for Efficient Network Management. In IWAN'99, Berlin, May 1999.

No context found.

D. Raz and Y. Shavitt. An active network approach for efficient network management. In Proceedings of the 1st International Working Conference on Active Networks (IWAN), pages 220 --231, 1999.

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