The improvement of file access performance is a great challenge in real-time cloud services. In this paper, we analyze preconditions of dealing with this problem considering the aspects of requirements, hardware, soft...
详细信息
ISBN:
(纸本)9780769548159
The improvement of file access performance is a great challenge in real-time cloud services. In this paper, we analyze preconditions of dealing with this problem considering the aspects of requirements, hardware, software, and network environments in the cloud. Then we describe the design and implementation of a novel distributed layered cachesystem built on the top of the Hadoop distributed File system which is named HDFS-based distributed cache system (HDcache). The cachesystem consists of a client library and multiple cache services. The cache services are designed with three access layers an in-memory cache, a snapshot of the local disk, and the actual disk view as provided by HDFS. The files loading from HDFS are cached in the shared memory which can be directly accessed by a client library. Multiple applications integrated with a client library can access a cache service simultaneously. cache services are organized in the P2P style using a distributed hash table. Every file cached has three replicas in different cache service nodes in order to improve robustness and alleviates the workload. Experimental results show that the novel cachesystem can store files with a wide range in their sizes and has the access performance in a millisecond level in highly concurrent environments.
In this letter, we propose a novel two-layered structured overlay, called a distributed Bloom filter table (DBFT), which is capable of substantially reducing overhead imposed by highly-dynamic peer-to-peer (P2P) distr...
详细信息
In this letter, we propose a novel two-layered structured overlay, called a distributed Bloom filter table (DBFT), which is capable of substantially reducing overhead imposed by highly-dynamic peer-to-peer (P2P) distributed storage systems. More specifically, in our DBFT scheme, each node's ID remains unchanged, when the associated cluster changes due to the physical movement of the node. This provides us with an exclusive benefit of maintaining the rule of structured overlay without imposing any additional overhead to reconfigure index between saved information and its stored node, which cannot be achieved by the previous P2P techniques designed for mobile ad hoc networks. Moreover, another merit of the proposed scheme is that stored packets per node become more equally distributed than in the conventional schemes, because each node maintains to have its own addressing-and content-data when it moves within a storage area.
distributedcache at mobile user devices and small base stations (SBSs) can improve the efficiency and reduce the delay for content dissemination in heterogeneous cellular networks (HetNets). One major challenge lies ...
详细信息
ISBN:
(纸本)9781509045884
distributedcache at mobile user devices and small base stations (SBSs) can improve the efficiency and reduce the delay for content dissemination in heterogeneous cellular networks (HetNets). One major challenge lies in the limited storage capacity of the participating mobile devices and SBSs. This paper investigates the design of an effective distributed content cache policy with heterogeneity (DCP-H) based on erasure code in terms of SBSs and mobile devices by considering sociality characteristics and physical communication conditions. The key feature of our proposed scheme lies in the PageRank algorithm based design, in which the transmit power and delivery delay is optimized through optimizing the cache states of content items to maximize cache hit ratio. Numerical results demonstrate the performance and advantage of our proposed scheme.
暂无评论