K. C. Almeroth, M. H. Ammar, and Z. Fei, "Scalable delivery of web pages using cyclic best-effort multicast," in Proceedings of INFOCOM, 1998, pp. 1214--1221.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....or a stock market quote is updated frequently, then caching of the document may be of little benefit, and may in fact actually increase latency. An alternative way of dealing with popular and frequently changing documents is to distribute them with continuous multicast push (CMP) 19] [1]. Here, a server housing a popular and frequently changing document continuously multicasts the latest version of the document on a multicast address. Clients tune into the multicast group for the time required to reliably receive the document and then leave the multicast group. More specifically, ....

K. V. Almeroth, M. H. Ammar, and Z. Fei, "Scalable Delivery of Web Pages Using Cyclic Best-Effort (UDP) Multicast", In Proceedings of IEEE INFOCOM, San Francisco, CA, USA, March 1998.

....differences play an important role. For example, while the no cache version is NP hard even for the unit file size case, our problem can be solved exactly in polynomial time in the case of unit size files. Several system designs also consider the use of broadcast or multicast to deliver files. [26] employs IP multicast to increase the capacity of a single web server. 27] and [28] explore a design in which a sender periodically transmits objects from a fixed library across a broadcast channel. In this system, the periodicity of a particular object is governed by its popularity. Other work ....

K. C. Almeroth, M. H. Ammar, and Z. Fei, "Scalable delivery of web pages using cyclic best-effort multicast," in Proceedings of IEEE INFOCOM'98, San Francisco, CA, March 1998.

....the same data is available, and download the data redundantly in parallel from all the sources. This may be done by applying some coding, e.g. Tornado Codes or Erasure Codes [5,6,7,8] Transmitting a file repeatedly to multicast groups that may be joined by any client that may need the file [5,8,9] takes advantage of the redundancy in sources and seems to be very efficient in the multi destination case (many to many) In the single destination case (many toone) it may result in some overhead in the total amount of transmitted data. This in turn may result in greater delay due to ....

K. C. Almeroth, M. H. Ammar, Z. Fei, "Scalable Delivery of Web Pages Using Cyclic Best-Effort Multicast," INFOCOM'98, pp. 1214-1221.

.... number of health officials and other authorized personnel when outbreaks of diseases are detected [40] Interestingly, RODS has the functionality of a Subscription Publisher server and exhibits all the requirements of a highly scalable Web server as described in the work of Almeroth et al. [5]. The objective of the Web server application is to scale to a large user (client) population, and scalability will be accomplished by using the multicast based middleware. In the middleware, the server can disseminate data by choosing any combination of the following three schemes: multicast ....

....improve user perceived performance, that is, it is common for different users to make requests to the same resource. Multicast pull is a good fit for warm resources for which repetitive multicast push cannot be justified, while there is an advantage in aggregating concurrent client requests [5]. Traditional unicast pull is reserved for cold documents. The end user should not perceive that Web resources are downloaded with a variety of methods, as the browser and the middleware shield the user from the details of the multitier dissemination protocol. In the RODS example as well as in a ....

[Article contains additional citation context not shown here]

K. C. Almeroth, M. H. Ammar, and Z. Fei. Scalable delivery of Web pages using cyclic best-effort (UDP) multicast. In Proc. of the Seventeenth Annual Joint Conference of the IEEE Computer and Communications Societies (INFOCOM 1998.

....a large number of health officials and other authorized personnel when outbreaks of diseases are detected . Interestingly, RODS has the functionality of a Subscription Publisher server and exhibits all the requirements of a highly scalable Web server as described in the work of Almeroth et al. [5]. The objective of the Web server application is to scale to a large user (client) population, and scalability will be accomplished by using the multicast based middleware. In the middleware, the server can disseminate data by choosing any combination of the following three schemes: multicast ....

....improve user perceived performance, that is, it is common for different users to make requests to the same resource. Multicast pull is a good fit for warm resources for which repetitive multicast push cannot be justified, while there is an advantage in aggregating concurrent client requests [5]. Traditional unicast pull is reserved for cold documents. The end user should not perceive that Web resources are downloaded with a variety of methods, as the browser and the middleware shield the user from the details of the multi tier dissemination protocol. In the RODS system as well as in a ....

[Article contains additional citation context not shown here]

K. C. Almeroth, M. H. Ammar, and Z. Fei. Scalable delivery of Web pages using cyclic best-effort (UDP) multicast. In Proceedings of the Seventeenth Annual Joint Conference of the IEEE Computer and Communications Societies (INFOCOM 1998.

....and optical networks. As a result, it is natural to develop broadcast applications for those media. Broadcast can also be used in networks other than wireless and optical as a method to solve scalability problems. For example, multicast methods can relieve the scalability problems of Web hot spots [1] and can support the operations of a content delivery network [2] Multicast methods can be combined with other performance enhancing techniques, such as caching [3] 4] Broadcast and multicast techniques have spawned research (e.g. 1] 5] and commercial ventures [6] 7] 8] that aim at ....

.... can relieve the scalability problems of Web hot spots [1] and can support the operations of a content delivery network [2] Multicast methods can be combined with other performance enhancing techniques, such as caching [3] 4] Broadcast and multicast techniques have spawned research (e.g. [1], 5] and commercial ventures [6] 7] 8] that aim at higher scalability. A common data dissemination method is to use repetitive server initiated multicast [9] 5] 3] whereby a server cyclically multicasts (or broadcasts) data to a large client population. As a general data management ....

[Article contains additional citation context not shown here]

K. C. Almeroth, M. H. Ammar, and Z. Fei, "Scalable delivery of Web pages using cyclic best-effort (UDP) multicast," in Proceedings of the Seventeenth Annual Joint Conference of the IEEE Computer and Communications Societies (INFOCOM 1998.

....deals with the efficient delivery of information to multiple users in overly crowded internet environment. Optimization of data flow between users and servers, can be achieved by intelligent caching, and multicasting schemes. The most recent work in this field has been done by Ammar et al. 6] [7]) On top of a reliable multicast data transfer protocol, they developed an application level data delivery policy, which partitions the web pages at the server side into hot, warm, and cold pages according to their popularity. In their proposed architecture, user requests reach the server via an ....

....connection. If it is for a warm page, then all of the requests for this page are grouped together, and serviced by a single multicast connection [6] The hot pages are multicast at a known multicast address periodically, and continuously, as soon as a request for them is received by the server [7]. The work described above, only dealt with the effective and efficient delivery of the information. Caching at both intermediate nodes and clients, may also reduce both the overall network traffic and the user. This problem has been considered in [8] 9] 10] Bestavros, in [8] introduced an ....

Almeroth, K. C., Ammar, M. H., Fei, Z., "Scalable Delivery of Web Pages Using Cyclic Best-Effort (UDP) Multicast," to appear in Proceedings of INFOCOM 98.

....method for delivering popular files with a minimal footprint on a single server leverages network mechanisms such as native multicast. Researchers have built scalable solutions which use non adaptive, cyclic transmissions over multicast or broadcast channels to provide eventual reliability [2], 1] and more sophisticated solutions which employ forward error correction to achieve scalable reliability without a significant performance penalty [12] 26] These solutions scale to large audiences, as the marginal cost of adding an additional client as perceived by the server is near ....

K. C. Almeroth, M. H. Ammar, and Z. Fei. Scalable delivery of web pages using cyclic best-effort (UDP) multicast. In Proceedings of IEEE INFOCOM, March 1998.

....[20] does not make a convincing case for splitting the traffic. We base our experiments on a production Apache server and we include persistent connections with pipelining in our experiments. Previously, Almeroth et al. proposed to use a UDP multicast for delivery of the most popular Web pages [21]. Using multicast to deliver popular Web pages to proxies has been proposed by Touch [22] In contrast to these works, we propose to use UDP for much of routine Web traffic. Analytical models for HTTP performance over TCP and ARDP, an alternative connection oriented transport protocol built over ....

K. C. Almeroth, M. H. Ammar, and Z. Fei, "Scalable delivery of web pages using cyclic best effort (udp) multicast," in INFOCOM, 1998.

....method for delivering popular files with a minimal footprint on a single server leverages network mechanisms such as native multicast. Researchers have built scalable solutions which use non adaptive, cyclic transmissions over multicast or broadcast channels to provide eventual reliability [2], 1] and more sophisticated solutions which employ forward error correction to achieve scalable reliability without a significant performance penalty [12] 26] These solutions scale to large audiences, as the marginal cost of adding an additional client as perceived by the server is near ....

K. C. Almeroth, M. H. Ammar, and Z. Fei. Scalable delivery of web pages using cyclic best-effort (UDP) multicast. In Proceedings of IEEE INFOCOM, March 1998.

....was first introduced in [10] In their architecture, clients use http requests. Several simultaneous requests are assembled, and grouped into a multicast group. Then, the clients join the dynamically generated multicast group and receive the page using a reliable multicast protocol. A later work [11] suggested the use of cyclic multicast over UDP for the same architecture, while determining the amount of time the server has to cyclicly transmit the same page until all users in the group receive it. The disadvantage of using this mechanism lies in the unnecessary overhead compared with our ....

Z. Fei K.C. Almeroth, M. Ammar, "Scalable delivery of web pages using cyclic-best-effor (udp) multicast," Infocom'98, Mar. 1998.

....[9] does not make a convincing case for splitting the traffic. We base our experiments on a production Apache server and we include persistent connections with pipelining in our experiments. Previously, Almeroth et al. proposed to use a UDP multicast for delivery of the most popular Web pages [2]. Using multicast to deliver popular Web pages to proxies has been proposed by Touch [33] In contrast to these works, we propose to use UDP for much of routine Web traffic. Analytical models for HTTP performance over TCP and ARDP, an alternative connection oriented transport protocol built over ....

K. C. Almeroth, M. H. Ammar, and Z. Fei. Scalable delivery of web pages using cyclic best effort (udp) multicast. In INFOCOM, 1998.

....method for delivering popular files with a minimal footprint on a single server leverages network mechanisms such as native multicast. Researchers have built scalable solutions which use non adaptive, cyclic transmissions over multicast or broadcast channels to provide eventual reliability [2], 1] and more sophisticated solutions which employ forward error correction to achieve scalable reliability without a significant performance penalty [13] 34] These solutions scale to large audiences, as the marginal cost of adding an additional client as perceived by the server is near ....

K. C. Almeroth, M. H. Ammar, and Z. Fei. Scalable delivery of web pages using cyclic best-effort (UDP) multicast. In Proceedings of IEEE INFOCOM, March 1998.

....removing system bottlenecks by reducing the load on a single server. For example, Almeroth, Ammar, and Fei suggest that servers holding heavily accessed web pages can reduce load and improve delivery performance by servicing requests in batches, rather than with separate point to point connections [2]. Their work advocates queuing popular requests at the server and establishing a multicast session over which to service the request once the queue length reaches a threshold. For pages which are in continuous demand, the use of reliable multicast with erasure codes may be an even better ....

K. C. Almeroth, M. Ammar, and Z. Fei, "Scalable Delivery of Web Pages Using Cyclic Best-Effort (UDP) Multicast," In Proc. of IEEE INFOCOM '98, San Francisco, 1998.

....or a stock market quote is updated frequently, then caching of the document may be of little benefit, and may in fact actually increase latency. An alternative way of dealing with popular and frequently changing documents is to distribute them with continuous multicast push (CMP) 19] [1]. Here, a server housing a popular and frequently changing document continuously multicasts the latest version of the document on a multicast address. Clients tune into the multicast group for the time required to reliably receive the document and then leave the multicast group. More specifically, ....

K. V. Almeroth, M. H. Ammar, and Z. Fei, "Scalable Delivery of Web Pages Using Cyclic Best-Effort (UDP) Multicast", In Proceedings of IEEE INFOCOM, San Francisco, CA, USA, March 1998.

....networks may be highly asymmetric (such as in satellite based networks) and because the system needs to scale to many receivers. Previous multicast work is also relevant since it demonstrates another approach for removing system bottlenecks by reducing the load on a single server. For example, [2] suggests that servers holding heavily accessed web pages can reduce load and improve delivery performance by servicing requests in batches, rather than with separate pointto point connections. Their work advocates queuing popular requests at the server and establishing a multicast session over ....

K. C. Almeroth, M. Ammar, and Z. Fei, "Scalable Delivery of Web Pages Using Cyclic Best-Effort (UDP) Multicast," In IEEE INFOCOM '98, San Francisco, 1998.

....to other materialized views. This work does not 77 address the issue of multiple differential update files, as required by our approach. Related work in aggregation of data for broadcast has been performed in many areas. This approach has been used to reduce bandwidth consumption of web servers[13, 42]. In [165] the authors analyzed data aggregation techniques for broadcasting data in Videotex systems. Our approach of dissemination of updates from server to the clients is related to work in broadcast disks[1, 2, 3, 4, 5, 6, 167] The motivation for this related work is to operate in an ....

Kevin C. Almeroth, Mostafa H. Ammar, and Zongming Fei. Scalable delivery of web pages using cyclic best-effort (UDP) multicast. In INFOCOM'98, May 1998.

....the system in real time. A number of projects have looked at broadcast of large information bases[18, 19, 20] The Broadcast Disks project[21] looks at disseminating data by cyclically broadcasting objects to clients. A similar project looked at cyclic, multicast delivery of web pages to clients[22]. Clients listen to a broadcast channel until they receive all subpieces of the informa tion they are interested in. An extension to the Broadcast Disks project investigates the balance between push versus pull data delivery[23] allowing clients to explicitly request information rather than ....

K. Almeroth, M. Ammar, and Z. Fei, "Scalable delivery of web pages using cyclic best-effort (udp) multicast," in Proceedings of IEEE INFOCOM 1998, (San Francisco, CA, USA), June 1998.

....useful in large scale systems that stream content to users. While video is a prime example, it is by no means the only content that can be streamed. Information dissementation and software distribution have recently become interesting types of content that can take advantage of similar techniques[12]. In comparing the efficiency and effectiveness of our technique to others, we choose to focus on three issues of particular relevance to the operation of large scale systems. These are: 2 ffl Service Characteristics: Content that is streamed is likely to require a set of server and delivery ....

....based resource allocation. 25 Another example is the dynamic allocation of multicast addresses, especially for applications that use many multicast addresses for long periods of time. A specific example, is the delivery of web pages using multicast and the cyclic, redundant transmission of pages[12]. Once a multicast address is allocated, it may be in use for a long period. If addresses for a particular application are allocated from a small, fixed range, large scale and long term cycles may develop. 6 Summary In this paper, we identify through simulation the limitations of greedy channel ....

K. Almeroth, M. Ammar, and Z. Fei, "Scalable delivery of web pages using cyclic best-effort (UDP) multicast," in IEEE Infocom, (San Francisco, California, USA), March 1998.

....of the system in real time. A number of projects have looked at broadcast of large information bases[4, 14, 16] The Broadcast Disks project[2] looks at disseminating data by cyclically broadcasting objects to clients. A similar project looked at cyclic, multicast delivery of web pages to clients[8]. Clients listen to a broadcast channel until they receive all subpieces of the information they are interseted in. An extension to the Broadcast Disks project investigates the balance between push versus pull data delivery[1] allowing clients to explicitly request information rather than always ....

K. Almeroth, M. Ammar, and Z. Fei. Scalable delivery of web pages using cyclic best-effort (udp) multicast. In Proceedings of IEEE INFOCOM 1998, San Francisco, CA, USA, June 1998.

No context found.

K. C. Almeroth, M. Ammar, and Z. Fei, "Scalable delivery of web pages using cyclic besteffort (UDP) multicast," in Proceedings of INFOCOM'98, 1998. San Francisco, California.

No context found.

K. C. Almeroth, M. H. Ammar, and Z. Fei, "Scalable delivery of web pages using cyclic best-effort multicast," in Proceedings of INFOCOM, 1998, pp. 1214--1221.

No context found.

Almeroth, K. C., Ammar, and M. H., Fei, Z. Scalable delivery of web pages using cyclic best-effort (UDP) multicast. Proceedings of INFOCOM, pages pp. 1214--1221, March 1998. 3

No context found.

K. C. Almeroth, M. H. Ammar, and Z. Fei, "Scalable delivery of web pages using cyclic best-effort multicast," in Proceedings of INFOCOM, 1998, pp. 1214--1221.

No context found.

K. Almeroth, M. Ammar, and Z. Fei. Scalable delivery of web pages using cyclic-besteffor (udp) multicast. Infocom'98, Mar. 1998.

First 50 documents

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