To solve the poor performance of existing cloud platforms in high-performance computing, our research team developed a high-performance cloud computing platform based on MPI. As a cloud platform, massive data storage ...
详细信息
ISBN:
(纸本)9781479941698
To solve the poor performance of existing cloud platforms in high-performance computing, our research team developed a high-performance cloud computing platform based on MPI. As a cloud platform, massive data storage is a key technology. In the research, we mainly studied how to build a MPI cloud platform that supports massive data storage, and implement a distributed cluster constructed by MySQL. This paper introduces the feature of MPI cloud platform, how to build the cloud platform database and the implementation of related SQL operations on the database. Based on this, we test the related operations in the database, and elaborate the shortcomings and challenges of the distributed database in this cloud computing platform.
As multicore processor architectures are now prevalent in server nodes of parallel and distributedcomputing systems, it has become important to characterize the performance of applications run on these architectures....
详细信息
ISBN:
(纸本)9781467376853
As multicore processor architectures are now prevalent in server nodes of parallel and distributedcomputing systems, it has become important to characterize the performance of applications run on these architectures. This study investigates the performance degradation an application experiences from memory interference due to other applications colocated on cores of the same multicore processor. We propose a methodology for designing models that are capable of utilizing varying amounts of information relating to an application and its co-located applications to predict the application's execution time performance degradation due to co-location. We evaluate the models sing several application co-location scenarios based on real world test data from two scientific benchmark suites on two server class Intel Xeon multicore processors.
Twig query is considered the core query pattern in most XML query language. With the XML document size becoming larger, single site cannot deal with such volume data in storage capacity and compute ability. Partitioni...
详细信息
ISBN:
(纸本)9781479941698
Twig query is considered the core query pattern in most XML query language. With the XML document size becoming larger, single site cannot deal with such volume data in storage capacity and compute ability. Partitioning the large data and distributed parallel processing query is an efficient and effective way. This paper proposes TwigMRR algorithm for evaluating XML twig query over large XML data that is encoded by Dewey, partitioned horizontally and distributed storage in a cluster. TwigMRR is based on MapReduce and extended a new model Map-Reduce-Reduce to get the final results for twig query. The experimental results show that our approach is scalable and efficient on this problem.
In conventional sensor networks, information sensed by distributed sensor nodes is collected to a sink node, and applications are executed based on this information. However, when actuator nodes in a field execute the...
详细信息
In conventional sensor networks, information sensed by distributed sensor nodes is collected to a sink node, and applications are executed based on this information. However, when actuator nodes in a field execute the applications, the sink node does not have to collect all the sensing information, but each actuator node collects the information required for its application. We aim to collect the required information to each actuator node, and generate a group whose members are source sensor nodes involved by each actuator node. We propose a protocol to efficiently collect sensing information of grouped source nodes. Furthermore, we experimentally demonstrate that actuator nodes can collect the information and that the proposed protocol can decrease the number of packets transmitted in the network.
Based on system dynamics modeling method, the nonlinear model of the cloud computing virtual machines scheduling is built. With system dynamics methodology and the use of simulation tools, from a system perspective on...
详细信息
ISBN:
(纸本)9781479941698
Based on system dynamics modeling method, the nonlinear model of the cloud computing virtual machines scheduling is built. With system dynamics methodology and the use of simulation tools, from a system perspective on cloud computing service requirements and building scale, the key factors affecting the number of cloud computing machines are identified. Furthermore, mathematical statistical theory and methods are used to find out optimal estimation and implementation of the construction scale with regarding to the decisive factor. The regression equations are submitted and the credibility of the regression equations is verified.
The advent of both Cloud computing and Internet of Things (IoT) is changing the way to conceive distributed systems. Nowadays, we can talk about IoT Cloud to indicate a new type of distributed system consisting of a s...
详细信息
The advent of both Cloud computing and Internet of Things (IoT) is changing the way to conceive distributed systems. Nowadays, we can talk about IoT Cloud to indicate a new type of distributed system consisting of a set of smart IoT devices or sensors interconnected with a remote Cloud infrastructure, platform, or software. Energy sustainability in IoT Cloud providers offers new tempting business opportunities for organizations, but at the same time it raises new challenges. In this paper, a flexible IoT Cloud federation energy management strategy is presented for optimizing the allocation of geographically localized smart sensors. We define the concept of IoT Cloud Federation as a mesh of IoT Cloud providers that are interconnected to provide a universal decentralized sensing and actuating environment where everything is driven by constraints and agreements in a ubiquitous infrastructure. In particular, a dynamic algorithm able to improve energy sustainability in a federated IoT Cloud ecosystem is discussed. In addition, we analyze a use-case driven strategy that allows both IoT Cloud providers and brokers to determine the paths to reach possible destination IoT devices in which computational resources should be dynamically migrated in order to push down the energy consumption due to IoT distributedapplications.
This paper covers the fast solution of large acoustic problems on low-resources parallel platforms. A domain decomposition method is coupled with a dynamic load balancing scheme to efficiently accelerate a geometrical...
详细信息
ISBN:
(纸本)9781479941698
This paper covers the fast solution of large acoustic problems on low-resources parallel platforms. A domain decomposition method is coupled with a dynamic load balancing scheme to efficiently accelerate a geometrical acoustic method. The geometrical method studied implements a beam-tracing method where intersections are handled as in a ray-tracing method. Beyond the distribution of the global processing upon multiple sub-domains, a second parallelization level is operated by means of multi-threading and shared memory mechanisms. Numerical experiments show that this method allows to handle large scale open domains for parallel computing purposes on few machines. Urban acoustic pollution arrising from car traffic was simulated on a large model of the Shinjuku district of Tokyo, Japan. The good speed-up results illustrate the performance of this new domain decomposition method.
Evaluation and performance analysis of an online collaborative project are never easy tasks because the massive human involvement and other qualitative factors are hard to assess. To figure out the relationship betwee...
详细信息
ISBN:
(纸本)9781479941698
Evaluation and performance analysis of an online collaborative project are never easy tasks because the massive human involvement and other qualitative factors are hard to assess. To figure out the relationship between human related factors and quality of collaboration outcomes, we propose an effective formal approach to estimate the human involvement in collaboration process and testify our method on 100 articles extracted from Wikipedia and Scholarpedia, the qualities of whose historical contents have been evaluated manually by volunteers of specific background. Through comparison of the human involvement and the outcome quality in these two projects, we find that the quality of collaborative products is positively related with the number, dedication and experience of collaboration participants, among which experience decides the necessary amount of human resources for a high quality, and increasing number or dedication of participants can also make up to the lack of experiences.
distributed key-value caching system has been deployed in many kinds of clouds. The effect of load balancing between each node is crucial in key-value caching system in clouds. Invalidation is a feature of the data in...
详细信息
ISBN:
(纸本)9781479941698
distributed key-value caching system has been deployed in many kinds of clouds. The effect of load balancing between each node is crucial in key-value caching system in clouds. Invalidation is a feature of the data in the key-value caching system, and making some cache invalid is efficient instead of adjusting the load location when load balancing. It is worth studying that how to utilize the feature reasonable to reach maximum load balancing. Aiming at this point, a new cache-invalidation-scope model in key-value caching system is proposed. Combined with greedy algorithm, the scheme provides a better load balancing algorithm for different load cases (CLB). CLB algorithm utilizes entropy and the scope of invalid cache invalid as the evaluation basis of load balancing effect. Compared with existing algorithm, CLB load balancing algorithm lifts up the performance of key-value caching system more than one times.
From the perspectives of the operational objectives of the green supply chain management and in consideration of its economic efficiency, social and environmental impact as well as its unique characteristics, this pap...
详细信息
ISBN:
(纸本)9781479941698
From the perspectives of the operational objectives of the green supply chain management and in consideration of its economic efficiency, social and environmental impact as well as its unique characteristics, this paper examines the pricing issues of the green supply chain for home appliances industry by using game theory. Considering the influences of the effective recycle behavior of the used home appliances on the whole supply chain, the paper proposes three game models for the portfolio pricing for the wholesale, retail and recycle price under three different collection methods respectively. The three collection methods are 1) manufacturer collection (Model M), 2) retailer collection (Model R) and 3) third-party collection (Model 3P). Then the paper analyzes how the wholesale price, the retail price for green home appliances, the recycle price for the used home appliances, and the total channel profits are affected by the choice of the collection methods. The pricing models presented in this paper provide a practical and theoretical guidance for home appliances enterprises in making pricing decisions.
暂无评论