Edge computing reduces connectivity costs and network traffic congestion over cloud computing, by offering local resources (processing and storage) at one hop closer to the end-users. I.e. it reduces the Round-Trip Ti...
详细信息
ISBN:
(纸本)9781728131016
Edge computing reduces connectivity costs and network traffic congestion over cloud computing, by offering local resources (processing and storage) at one hop closer to the end-users. I.e. it reduces the Round-Trip Time (RTT) for offloading part of the processing workload from end-nodes/devices to servers at the edge. However, edge servers are normally pre-setup as part of the overall computing resource infrastructure, which is tough to predict for mobile/IoT deployments. This paper introduces a smart Dynamic Edge Offloading scheme, (we named it DEO), that forms the "edge computing resource" on-the-go, as needed from nearby available devices in a cooperative sharing environment. This is especially necessary for hosting mobile/IoT applications traffic at crowded/urban situations, and, for example, when executing a processing intensive Mobile Cloud Computing Service (MCCS) on a Smartphone (SP). DEO implementation is achieved by using a short-range wireless connectivity between available cooperative end-devices, that will form the edge computing resource. DEO includes an intelligent cloud-based engine, that will facilitate the engagement of the edge network devices. For example, if the end-device is a SP running an MCCS, DEO will partition the processing of the MCCS into sub-tasks, that will be run in parallel on the newly formed "edge resource network" of other nearby devices. Our experiments prove that DEO reduces the RTT and cost overhead by 62.8% and 75.5%, when compared to offloading to a local edge server or a cloud-based server.
Mobile computation offloading of computational intensive tasks from mobile devices to surrogate cloud servers has been recently envisaged as a promising technique to enhance the computational capacity of the mobile de...
详细信息
ISBN:
(纸本)9781479927302
Mobile computation offloading of computational intensive tasks from mobile devices to surrogate cloud servers has been recently envisaged as a promising technique to enhance the computational capacity of the mobile devices. Within this framework we consider a MIMO multicell system wherein several Mobile Users (MUs) ask for computation offloading to a common cloud server through their femto-access points. We formulate the computation offloading problem as a joint optimization of the radio and computational resources in order to minimize the overall users' energy consumption while meeting the latency constraints imposed by the applications. To solve this non-convex problem we hinge on successive convex approximation techniques by showing that the original problem can be decomposed in parallel convex subproblems. Hence we devise an iterative algorithm which can be implemented in a distributed manner across the access points through dual/primal decomposition techniques requiring limited coordination/signaling with the cloud.
Scientific computing applications with highly demanding data capacity and computation power drive a computing platform migration from shared memory machines to multi-core/multiprocessor computer clusters. However, ove...
详细信息
ISBN:
(纸本)9783540680819
Scientific computing applications with highly demanding data capacity and computation power drive a computing platform migration from shared memory machines to multi-core/multiprocessor computer clusters. However, overheads in coordinating operations across computing nodes could counteract the benefit of having extra machines. Furthermore, the hidden dependency in applications slows down the simulation over non-shared memory machines. This paper proposed a framework to utilize multi-core/multiprocessor clusters for distributed simulation. Among several coordination schemes, decentralized control approach has demonstrated its effectiveness in reducing the communication overheads. A speculative execution strategy is applied to exploit parallelism thoroughly and overcome strong data dependency. Performance analysis and experiments are provided to demonstrate the performance gains.
This poster investigates sensory data processing, filtering and sensor fusion methods for autonomous vehicles operating in real-life, urban environments with human and machine drivers, and pedestrians. Extended Kalman...
详细信息
ISBN:
(纸本)9780769546957
This poster investigates sensory data processing, filtering and sensor fusion methods for autonomous vehicles operating in real-life, urban environments with human and machine drivers, and pedestrians. Extended Kalman Filters were used to develop decentralized data fusion algorithms for communicating vehicles;Particle Filters were improved by assigning trust/confidence values in order to overcome faulty/compromised sensors;and the computational cost of particle filters were distributed by parallelizing the load using the developed Shared-Memory Systematic Resampling algorithm.
Frequent Itemset Mining (FIM) is the core of tasks such as association rules and sequential pattern mining. With the increasing amount of data, traditional FIM algorithms become inefficient due to excessive resource r...
详细信息
ISBN:
(数字)9781728197418
ISBN:
(纸本)9781728197418
Frequent Itemset Mining (FIM) is the core of tasks such as association rules and sequential pattern mining. With the increasing amount of data, traditional FIM algorithms become inefficient due to excessive resource requirements or high communication costs. In this paper, the Eclat algorithm in the frequent itemset mining algorithm is taken as the research point, and the parallel Eclat optimization algorithm BPEclat (Balanced parallel Eclat) based on Spark is proposed to solve the performance shortcoming of Eclat algorithm in serial processing large-scale data. The algorithm is improved and optimized from many aspects: combining the pre-pruning and post-pruning depth pruning strategies to reduce the calculation of irrelevant itemset, compressing the candidate set size;using the prefix term to divide the data set, and using the range partitioning idea to balance the calculations node load, improve the parallel computing power of the algorithm. The experimental results show that the proposed BPEclat algorithm reduces the candidate set size by 25.3% and the time consumption by 32.5%. Therefore, it is possible to process massive amounts of data more efficiently and reliably, and has good scalability and universality.
In this paper, we describe KOSHIK, an end-to-end framework to process the unstructured natural language content of multilingual documents. We used the Hadoop distributed computing infrastructure to build this framewor...
详细信息
This paper describes the first results from research on the compilation of constraint systems into task level parallel programs in a procedural language. This is the only research of which we are aware which attempts ...
详细信息
This paper describes the first results from research on the compilation of constraint systems into task level parallel programs in a procedural language. This is the only research of which we are aware which attempts to generate efficient parallel programs for numerical computation from constraint systems. Computations are expressed as constraint systems. A dependence graph is derived from the constraint system and a set of input variables. The dependence graph, which exploits the parallelism in the constraints, is mapped to the language CODE, which represents parallel computation structures as generalized dependence graphs. Finally, parallel C programs are generated. To extract parallel programs of appropriate granularity, the following features are included: (i) modularity, (ii) operations over structured types as primitives, (iii) sequential C functions. A prototype of the compiler has been implemented. The domain of matrix computations is targeted for applications. Initial results are very encouraging.
This paper presents a case for exploiting the synergy of dedicated and opportunistic network resources in a distributed hosting platform for data stream processingapplications. Our previous studies have demonstrated ...
详细信息
ISBN:
(纸本)9780769534343
This paper presents a case for exploiting the synergy of dedicated and opportunistic network resources in a distributed hosting platform for data stream processingapplications. Our previous studies have demonstrated the benefits of combining dedicated reliable resources with opportunistic resources in case of high-throughput computing applications, where timely allocation of the processing units is the primary concern. Since distributed stream processingapplications demand large volume of data transmission between the processing sites at a consistent rate, adequate control over the network resources is important here to assure a steady flow of processing. In this paper, we propose a system model for the hybrid hosting platform where stream processing servers installed at distributed sites are interconnected with a combination of dedicated links and public Internet. Decentralized algorithms have been developed for allocation of the two classes of network resources among the competing tasks with an objective towards higher task throughput and better utilization of expensive dedicated resources. Results from extensive simulation study show that with proper management, systems exploiting the synergy of dedicated and opportunistic resources yield considerably higher task throughput and thus, higher return on investment over the systems solely using expensive dedicated resources.
distributed resources required for processing complex parallelapplications are becoming larger in scale day-by-day. They need to be used efficiently in order to provide Quality of Service (QoS). Therefore, resource a...
详细信息
ISBN:
(纸本)9781728116723
distributed resources required for processing complex parallelapplications are becoming larger in scale day-by-day. They need to be used efficiently in order to provide Quality of Service (QoS). Therefore, resource allocation and scheduling are of paramount importance. In this paper we investigate issues involved with the scheduling of parallel jobs that are bag-of-task-chains in distributed systems. Two task-chain scheduling techniques in two different cases of resource allocation are studied. The performance of the task-chain scheduling algorithms is evaluated via simulation, under different cases of system workload. The experimental results show that the performance of the task-chain scheduling strategies depends on the employed resource allocation techniques.
MrBayes is a popular bioinformatics software that is widely used in phylogenetic analysis. The core algorithm of Mrbayes is Metropolis Coupled Markov Chain Monte Carlo (MC3). However, when dealing with large data sets...
详细信息
ISBN:
(纸本)9781538637906
MrBayes is a popular bioinformatics software that is widely used in phylogenetic analysis. The core algorithm of Mrbayes is Metropolis Coupled Markov Chain Monte Carlo (MC3). However, when dealing with large data sets, MC3 algorithm is too slow to meet researcher's requirements. Although several parallelizations have been proposed for MrBayes, such as MPI (Message Passing Interface) based MrBayes, GPU (Graphics processing Unit) based MrBayes, there is still no efficient parallel algorithm to fully utilize computing power of modern CPU and computer architecture. This paper (a) presents a new three-level hybrid parallel algorithm, include data-level parallelism (DLP), thread-level parallelism (TLP), and process-level parallelism (PLP), which can be used on most modern multi-core computers;(b) compares the performance of different combinations of parallel strategies on real-world protein data sets. The experimental results show that, this hybrid parallel algorithm does convert more computing powers into higher speedup. Furthermore, the proposed algorithm's speedup is near the speedup on one GPU at the same data sets. This algorithm is fit for practical use in phylogenetic inferences.
暂无评论