In this paper the idea that large objects, such as video files, should not be cached or replaced in their entirety, but rather be partitioned in chunks and replacement decisions be applied at the chunk level is examined. It is shown, that a higher byte hit ratio (BHR) can be achieved through partial replacement. The price paid for the improved BHR performance is that the replacement algorithm, e.g. LRU, takes a longer time to induce the steady state BHR. It is demonstrated that this problem could be addressed by a hybrid caching scheme that employs variable sized chunks; the use of small chunks leads to the maximization of BHR in periods of stable video popularity, while large chunks are used when extreme popularity changes occur to assist the fast convergence to the new steady state BHR.
No abstract
Partial caching of large media objects such as video files has been proposed recently as the caching of entire objects can easily exhaust the storage resources of a proxy server. In this paper the idea of segmenting video files into chunks and applying replacement decisions at the chunk level rather than on entire videos is examined. It is shown that a higher byte hit ratio (BHR) can be achieved by appropriately adjusting the replacement granularity. The price paid for the improved BHR performance is that the replacement algorithm takes a longer time to converge to the steady state BHR. For the segmentation of video into chunks two methods are presented. The Fixed Chunk Size segmentation scheme that is rather simple and reveals the basic trade-off between byte hit ratio (BHR) and responsiveness to changes of popularity; the Variable Chunk Size segmentation scheme that uses the request frequencies to dynamically adjust the size of the chunk and is shown to be capable of combining a small response time with high BHR. Moreover, a variation of the fixed chunk size segmentation scheme is presented, which is shown to improve its performance by switching between different chunk sizes. Video segmentation is also considered as a mechanism to provide for caching differentiation based on access costs. By employing access cost dependent chunk sizes an overall access cost reduction is demonstrated.
Abstract-Partial caching of large media objects such as video 6les has been proposed in men1 pmxy caching sch-, as the caching of entire videas can exhaust proxy storage resources within few requests. In this work the idea of segmenting videos inlo a nnmber d chunks and replacement decisions be taken at the chunk level rather than based on the entire objects is adopted. In particular a video segmentation scheme called Variable Chunk Size is presented. This scheme uses the request frequenck to dynamically adjust the s u e of the chunk and it is shown to be capable of combining a high BHR with m a l l ".se time to demand changes. Video segmentation is also considered as a mechanism to provide for caching diller. entialion based on access costs. By employing access cast dependent chunk sizes an o v e d acces cast reduction is demonstrated.
Abstract. Recently, the application of the peer to peer networking paradigm (typical for an ad hoc network) has been proposed for wireless local area networks (WLANs), instead of the traditional cellular networking paradigm. In this paper the performance of a WLAN employing the peer to peer networking paradigm is studied via simulations; the results indicate that the direct application of the peer to peer networking paradigm in a WLAN leads to a substantially decreased throughput for the traffic directed to the Access Point (AP). The study also reveals that the cumulative receiving throughput of nodes located at the periphery of relatively small circular areas around the AP is substantially higher. Thus, the capacity of the multihop cellular network may be enhanced by employing the peer to peer paradigm only outside a circular area around the AP and the cellular paradigm inside this circular area. Examples are provided of environments where the aforementioned idea of distributing the traditional AP functionality to a set of nodes at the periphery of a circular area around the AP can be effectively applied.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.