Results 1 -
4 of
4
A Bayesian Game Formulation of Power Dissipation and Response Time Minimization in a Mobile Cloud Computing System
"... Abstract—The rapidly developing cloud computing and virtualization techniques provide mobile devices with battery energy saving opportunities by allowing them to offload computation and execute applications remotely. A mobile device should judiciously determine whether to offload computation and whi ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
(Show Context)
Abstract—The rapidly developing cloud computing and virtualization techniques provide mobile devices with battery energy saving opportunities by allowing them to offload computation and execute applications remotely. A mobile device should judiciously determine whether to offload computation and which portion of application should be offloaded to the cloud. This paper considers a mobile cloud computing (MCC) interaction system consisting of multiple mobile devices and the cloud computing system. A Bayesian game formulation is proposed for the MCC interaction system. In this game, each mobile device determines the portion of its service requests for remote processing in the cloud computing system. All the mobile devices compete for the allocated resources in the data center. Each mobile device is aware of its own service request generating rate through effective prediction methods. It has only partial information about the other mobile devices. The objective of each mobile device is to minimize its power consumption as well as the service request response time. This paper proves that pure strategy Bayesian-Nash equilibrium in this game always exists and is unique. The optimal strategy for all the mobile devices achieving the Bayesian-Nash equilibrium is derived using convex optimization technique. Experimental results demonstrate the effectiveness of the proposed Bayesian game-based optimization framework. The mobile devices can achieve simultaneous reduction in average power consumption and average service request response time, by 27.3 % and 63.7%, respectively, compared with baseline methods. Keywords-mobile cloud computing; mobile devices; game theory; Bayesian game; resource allocation I.
Evaluating Orthogonality between Application Auto-Tuning and Run-Time Resource Management for Adaptive OpenCL Applications
"... Abstract-The ever increasing number of processing units integrated on the same many-core chip delivers computational power that can exceed the performance requirements of a single application. The number of chips (and related power consumption) can thus be reduced to serve multiple applications -a ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract-The ever increasing number of processing units integrated on the same many-core chip delivers computational power that can exceed the performance requirements of a single application. The number of chips (and related power consumption) can thus be reduced to serve multiple applications -a practice which is called resource consolidation. However, this solution requires techniques to partition and assign resources among the applications and to manage unpredictable dynamic workloads. To provide the performance requirements in such scenarios, we exploit application auto-tuning, based on design-time analysis, of both application-specific dynamic knobs and computational parallelism. Such features are implemented in a software library, which is used to demonstrate the main contribution of this paper: a light-weight Run-Time Resource Management -RTRM -technique to improve resource sharing for computationally intensive OpenCL applications. We evaluate how much the interaction between RTRM and application auto-tuning can become synergistic yet orthogonal. In the proposed approach, run-time adaptation decisions are taken by each application, autonomously. This has two main advantages: i) a non-invasive application design, in terms of source code, and ii) a very low run-time overhead, since it does not require any central coordination of a supervisor nor communication between the applications. We carried out an experimental campaign by using a video processing application -an OpenCL stereo-matching implementation -and stressing out resource usage. We proved that, while RTRM is necessary to provide lower variance of the application performance, the application auto-tuning layer is fundamental to trade it off with respect to the computation accuracy.
A Nested Two Stage Game-Based Optimization Framework in Mobile Cloud Computing System
"... Abstract—The rapidly developing cloud computing and virtualization techniques provide mobile devices with battery energy saving opportunities by allowing them to offload computation and execute applications remotely. A mobile device should judiciously decide whether to offload computation and which ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract—The rapidly developing cloud computing and virtualization techniques provide mobile devices with battery energy saving opportunities by allowing them to offload computation and execute applications remotely. A mobile device should judiciously decide whether to offload computation and which portion of application should be offloaded to the cloud. In this paper, we consider a mobile cloud computing (MCC) interaction system consisting of multiple mobile devices and the cloud computing facilities. We provide a nested two stage game formulation for the MCC interaction system. In the first stage, each mobile device determines the portion of its service requests for remote processing in the cloud. In the second stage, the cloud computing facilities allocate a portion of its total resources for service request processing depending on the request arrival rate from all the mobile devices. The objective of each mobile device is to minimize its power consumption as well as the service request response time. The objective of the cloud computing controller is to maximize its own profit. Based on the backward induction principle, we derive the optimal or near-optimal strategy for all the mobile devices as well as the cloud computing controller in the nested two stage game using convex optimization technique. Experimental results demonstrate the effectiveness of the proposed nested two stage game-based optimization framework on the MCC interaction system. The mobile devices can achieve simultaneous reduction in average power consumption and average service request response time, by 21.8 % and 31.9%, respectively, compared with baseline methods. Keywords-mobile cloud computing; mobile devices; game theory; nested game; resource allocation I.
2013 IEEE Green Technologies Conference Service Level Agreement-Based Joint Application Environment Assignment and Resource Allocation in Cloud Computing Systems
"... Abstract—Cloud computing have attracted a lot of attention recently due to increasing demand for high performance computing and storage. Resource allocation is one of the most important challenges in the cloud computing system especially when the clients have some Service Level Agreements (SLAs) and ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract—Cloud computing have attracted a lot of attention recently due to increasing demand for high performance computing and storage. Resource allocation is one of the most important challenges in the cloud computing system especially when the clients have some Service Level Agreements (SLAs) and the total profit depends on how the system can meet these SLAs. Moreover, a data center typically hosts and manages a suite of application environments and a fixed number of servers that are allocated to these application environments in a way that maximizes a certain utility function. In this paper, we consider the problem of SLA-based joint optimization of application environment assignment, request dispatching from the clients to the servers, as well as resource allocation in a data center comprised of heterogeneous servers. The objective is to maximize the total profit, which is the total price gained from serving the clients subtracted by the operation cost of the data center. The total price depends on the average service request response time for each client as defined in their utility functions, while the operating cost is related to the total energy consumption. We propose a near-optimal solution of the joint optimization problem based on the Hungarian algorithm for the assignment problem, as well as convex optimization techniques, in a way that is similar to the constructive partitioning algorithm in VLSI computer-aided design (CAD). Experimental results demonstrate that the proposed nearoptimal joint application environment assignment and resource allocation algorithm outperforms baseline algorithms by up to 65.7%. Keywords-cloud computing; application environment; resource allocation; assignment problem I.
