Modern GPGPUs (General-Purpose Graphics Processing Units) have the ability of executing thousands of threads simultaneously. However, the resource utilization of GPGPU in real systems is limited as the load balancing ...
详细信息
ISBN:
(纸本)9781665495530
Modern GPGPUs (General-Purpose Graphics Processing Units) have the ability of executing thousands of threads simultaneously. However, the resource utilization of GPGPU in real systems is limited as the load balancing between SMs (Stream Multiprocessors) is difficult during the scheduling of thread blocks, which are the basic units for resource allocation in GPGPU. In order to schedule thread blocks in GPGPU, the current hardware scheduler allocates thread blocks to SMs by the Round-Robin order. Although this is simple and easy to implement, we show that Round-Robin is not efficient when thread blocks of heterogeneous workloads are mixed. In such environments, efficient resource sharing in GPGPU is challenging as workloads have different resource usage patterns, but scheduling should be performed instantly. In this paper, we present a new thread block scheduling algorithm that has the ability of analyzing the load of SMs and the characteristics of pending thread blocks. Specifically, we formulate thread block scheduling as a bin-packing problem, and aim to minimize the internal fragmentation of SMs by arranging size-aware filling of thread blocks to overall SMs in advance. To do so, we make use of multiple queues for incoming thread blocks according to their sizes and perform scheduling by considering the load balancing of SMs. Our experimental results under a wide range of workload conditions show that the proposed algorithm improves the performance of GPGPU by 24.8% on average compared to the Round-Robin scheduler.
For great change of service time for request,big difference of hardware and software server and different network performance,this paper proposes a dynamic-feedback algorithm based on AHP in the course of studying the...
详细信息
For great change of service time for request,big difference of hardware and software server and different network performance,this paper proposes a dynamic-feedback algorithm based on AHP in the course of studying the algorithm of load balancing in the cluster-based system. Combined with Weighted scheduling algorithm of the kernel, based on the parameters influencing the performance of cluster system from dynamic feedback,we can adjust the servers'weight,solve the load imbalance problem among the servers effectively and certainly improve the throughput of the whole system.
A significant amount of energy is consumed to render high-level computation tasks in large scale cloud computing applications. The state-of-the-art energy saving techniques based on centralized job placement approache...
详细信息
ISBN:
(纸本)9781479909735
A significant amount of energy is consumed to render high-level computation tasks in large scale cloud computing applications. The state-of-the-art energy saving techniques based on centralized job placement approaches reduce the reliability of operation due to a single point of failure. Moreover, the existing works do not consider energy consumption cost for communication devices and network appliances which contribute a lot. In this paper, we have proposed a mechanism for cluster formation based on network vicinity among the data servers. We have developed two distributed and localized intra-cluster and inter-cluster VM scheduling algorithms based on energy calculation, resource requirement and availability. Our proposed scheduling algorithms manage VMs to reduce the energy consumption of both the servers and networking devices. Simulation results show that our proposed distributed VM scheduling algorithms can conserve significant amount of energy compared to state-of-the-art works.
The cloud based innovative applications are increasing regularly and hence the data and job load also increasing proportionally. Cloud based service providers are also increasing their infrastructure and service facil...
详细信息
ISBN:
(纸本)9781665439718
The cloud based innovative applications are increasing regularly and hence the data and job load also increasing proportionally. Cloud based service providers are also increasing their infrastructure and service facility to serve in a much better way to its clients. The job processing load will also increase the waiting time and hence affect the service response time at user’s end. So, it is always a matter of great importance that which job scheduling algorithm should be applied to serve the client in an efficient manner. This is the main motivation for framing this research paper. In this paper, we are taking the main five heuristic job scheduling algorithms like FCFS (First Come First Server), SJF (Shortest Job First), MaxMin, MinMin, and Saffrage for analyzing on the pre-decided cloud infrastructure. Among these heuristic algorithm, MaxMin algorithm outperforms than others in all the test cases i.e. with the cloudlet load of 100, 200, 300, …, 1000 cloudlets. Hence we can say that the MaxMin is the best scheduling algorithm among these five heuristic job scheduling algorithms.
Cloud Computing is a recent developmental paradigm in the field of computing offering huge power to next generation computers. The dynamic provisioning acts as a base for cloud computing facilitating and supporting th...
详细信息
Cloud Computing is a recent developmental paradigm in the field of computing offering huge power to next generation computers. The dynamic provisioning acts as a base for cloud computing facilitating and supporting the network services. It focuses on making the vision of utility computing a reality with pay-as-you-go. It offers immense potential to bloom the world with applications and products focussing on greater resource utilization and scalability. This paper presents the basic cloud computing fundamentals and the concepts of load balancing i.e., scheduling of load in the cloud. It elaborates the existing load scheduling algorithms with their merits/demerits and suitability in the cloud and heterogeneous computing environment and proposes a new perspective for better results as per desired parameters.
Efficient workflow scheduling algorithms are crucial for attaining high performance in large-scale heterogeneous distributed infrastructures, such as cloud computing. List scheduling algorithms are one of the most eff...
详细信息
ISBN:
(数字)9781728160955
ISBN:
(纸本)9781728196497
Efficient workflow scheduling algorithms are crucial for attaining high performance in large-scale heterogeneous distributed infrastructures, such as cloud computing. List scheduling algorithms are one of the most efficient heuristic methods for assigning task graphs to fully connected heterogeneous systems. However, most existing list-based scheduling algorithms do not consider the applications' security requirements and the security services offered by cloud providers. In this paper, we extend four list scheduling algorithms for security-aware workflow scheduling in the IaaS cloud. The idea of the extension is to consider the security overheads in both tasks prioritizing phase and virtual machine selection phase of the four original algorithms. Based on real-world applications, we evaluate the performance of the proposed algorithms in terms of scheduling length, speedup and efficiency.
Wireless mesh networks (WMNs) have been developed to answer the needs of many wireless applications. A major limiting parameter of the performance of WMNs is the interference between the several communications that oc...
详细信息
ISBN:
(纸本)9781479909582
Wireless mesh networks (WMNs) have been developed to answer the needs of many wireless applications. A major limiting parameter of the performance of WMNs is the interference between the several communications that occur simultaneously in the same network. To address this limitation, an adequate scheduling algorithm has to be implemented. Hence, this paper focuses on the scheduling problem under the physical interference model. This problem is known to be NP-Hard problem. In this paper we develop and propose two efficient scheduling algorithms. We evaluate their performances by simulation in terms of spatial reuse and we compare their performance with known previously-proposed algorithms. We show that our proposed algorithms provide high performances with low complexity.
This paper discusses the synthesis and implementation of various scheduling algorithms for Network-on-Chip communication. Traditionally these scheduling algorithms were implemented on ASIC platforms generally for shar...
详细信息
This paper discusses the synthesis and implementation of various scheduling algorithms for Network-on-Chip communication. Traditionally these scheduling algorithms were implemented on ASIC platforms generally for shared bus based interconnection systems. In this paper we carry a comparative analysis by synthesizing and implementing various scheduling algorithms for configuring the crossbar in input queued switches. The implementation is carried out using various arbitration networks responsible for scheduling 8-bit input requests. The implementation targets Spartan6 FPGA family. The analysis concludes that the scheduling algorithm based on CLA based encoding network shows lower power delay product and lower area delay product and a reasonably lower resource utilization when implemented for speed optimization goal.
In recent years, a suite of Glauber dynamics-based CSMA algorithms have attracted great attention due to their simple, distributed implementations with guaranteed throughput-optimality. However, these algorithms often...
详细信息
ISBN:
(纸本)9781467399548
In recent years, a suite of Glauber dynamics-based CSMA algorithms have attracted great attention due to their simple, distributed implementations with guaranteed throughput-optimality. However, these algorithms often suffer from poor delay performance and the starvation problem. Among several attempts to improve the delay performance, a remarkable improvement has recently been made in a class of CSMA algorithms that utilize multiple instances of the algorithm (or Markov chains). In this paper, we develop a new approach via an antithetic coupling (AC) method, which can further improve the delay performance of those that virtually emulate multiple chains. The key enabler of utilizing AC method lies in our skilful choice of manipulating the driving sequences of random variables that govern the evolution of schedule instances, in such a way that those multiple instances of chains become negatively correlated as oppose to having them run independently. This contributes faster change of the link state, rendering it more like a periodic process and thus leading to better queueing performance. We rigorously establish an ordering relationship for the effective bandwidth of each net-input process to the queue, between our proposed algorithm (AC-CSMA) and other state-of-the-art existing algorithms in the literature, under a mild set of assumptions. The proposed algorithm involves very simple modification onto existing CSMA-based algorithms, and can be implemented in a fully distributed manner without any additional message overhead. Our extensive simulation results also confirm that AC-CSMA always delivers better queueing performance over a variety of network scenarios.
暂无评论