Dynamic Switching-Frequency Scaling: Scheduling Overcommitted Domains in Xen VMM

Chen, Huacai; Jin, Hai; Hu, Kan; Huang, Jian

doi:10.1109/icpp.2010.36

Cited by 11 publications

(10 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some related work pertain to network virtualization techniques to fully exploit the benefits of virtualization is summarized in [12]. Other research with the same goal in the multicore environments includes [4], [10], [11]. However, no one in these studies considers all the problems we delved into in this paper.…”

Section: ) Mean Response Time Of Vcpusmentioning

confidence: 98%

“…RELATED WORK Improving the Credit scheduler to assure that I/O domains will get timely response has long been an active research topic. Although various methods have been proposed [4], [5], [8]- [11], in practice I/O latency is still an obstacle, leaving much room unexplored for further improvements. Govindan et al [8] proposed a communication-aware scheduling to preferentially scheduling communication-oriented domains over computeintensive counterparts by scarifying a short-term fairness.…”

Section: ) Mean Response Time Of Vcpusmentioning

confidence: 99%

“…Chen et al [4] discussed dynamic time slice problem based on their analysis on the ineffective holding time in the communication-intensive multiprocessing programs. Similar to ours, they also suggested a set of formulae to compute the variable time slice in order to dynamically scale the context switching frequency.…”

Section: ) Mean Response Time Of Vcpusmentioning

confidence: 99%

“…This would adversely impact the overall system performance as the control domain (Dom0) back-ends the communication directly with I/O devices to complete all the I/O requests in the system. On the other hand, the fixed time slice the Credit scheduler algorithm assigns to each PCPU may also potentially cause frequent switching, cache miss or low hit rate, and sporadic task starvation [4].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

An Improved Xen Credit Scheduler for I/O Latency-Sensitive Applications on Multicores

Zeng

Wang

Shi

et al. 2013

2013 International Conference on Cloud Computing and Big Data

View full text Add to dashboard Cite

It has long been recognized that the Credit scheduler favors CPU-bound applications while for the latency-sensitive workloads such as those related to stream-based audio/video services, its performance is far from satisfactory. In this paper we present an improved Credit scheduler in Xen to facilitate such tasks on multicore platforms. To this end, we improve the Credit scheduler from three perspectives. First, given the identified Simultaneous Multi-Boost problem, we minimize the system response time by load balancing the virtual CPUs with the BOOST priority between the cores. Second, we address the Premature Preemption problem by monitoring the received network packets in the driver domain and deliberately preventing it from being prematurely preempted during the packet delivery to further reduce and stabilize the I/O latency. Finally, we optimize the frequency of CPU switch by utilizing time-variant slice instead of the existing long time-invariant one to adapt to the dynamic fluctuation of the number of virtual CPUs in the run queue associated with each physical CPU. Our empirical studies show that the proposed improvement can significantly improve the performance of the Credit scheduler for scheduling the I/O latency-sensitive applications.

show abstract

Section: ) Mean Response Time Of Vcpusmentioning

confidence: 98%

Section: ) Mean Response Time Of Vcpusmentioning

confidence: 99%

Section: ) Mean Response Time Of Vcpusmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Improved Xen Credit Scheduler for I/O Latency-Sensitive Applications on Multicores

Zeng

Wang

Shi

et al. 2013

2013 International Conference on Cloud Computing and Big Data

View full text Add to dashboard Cite

show abstract

“…As the Credit algorithm is the most frequently used algorithm for the scheduler in Xen-based virtual machines, it has inspired great research interests in both academia and industry [5], [10], [11], [13], [19]. We will view some related work as per main research interests in the literature.…”

Section: Related Workmentioning

confidence: 99%

XCollOpts: A Novel Improvement of Network Virtualizations in Xen for I/O-Latency Sensitive Applications on Multicores

Zeng

Wang

Feng

et al. 2015

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

It has long been recognized that the Credit scheduler selectively favors CPU-bound applications whereas for I/O-latency sensitive workloads, such as those related to stream-based audio/video services, it only exhibits tolerable, or even worse, unacceptable performance. The reasons behind this phenomenon are the poor understanding (to some degree) of the virtual machine scheduling as well as the network I/O virtualizations. In order to address these problems and make the system more responsive to the I/O-latency sensitive applications, in this paper, we present XCollOpts which performs a collection of novel optimizations to improve the Credit scheduler and the underlying I/O virtualizations in multicore environments, each from two perspectives. To optimize the schedule, in XCollOpts, we first pinpoint the Imbalanced Multi-Boosting problem among the cores thereby minimizing the system response time by load balancing the BOOST VCPUs. Then, we describe the Premature Preemption problem and address it by monitoring the received network packets in the driver domain and deliberately preventing it from being prematurely preempted during the packet delivery. However, these optimizations on the scheduling strategies cannot be fully exploited if the performance issues of the underlying supportive communication mechanisms are not considered. To this end, we make two further optimizations for the network I/O virtualizations, namely, Multi-Tasklet Pairs and Optimized Small Data Packet. Our empirical studies show that with XCollOpts, we can significantly improve the performance of the latency-sensitive applications at a cost of relatively small system overhead.Index Terms-Scheduling, hypervisor, multicore, load balancing, I/O virtualization.

show abstract

An analysis of performance factors on Esper‐based stream big data processing in a virtualized environment

Choi

Dai

2014

Int J Communication

View full text Add to dashboard Cite

To process big data in a cloud computing environment, a large scale of stream big data is one of the challenging issues. In particular, complex event processing is an emerging technology to handle massive events from multiple sources and process them in a near real time. In this paper, we analyze four performanceinfluencing factors on a virtualized event processing system: the number of query statements, garbage collection intervals, the quantity of virtual machine resources, and virtual CPU assignment types. In our experiments, we observe the performance effects of these performance parameters, by implementing and running an Esper-based event processing application on top of a Xen-based virtualized system. With experimental results, we analyze the memory consumption problem, and apply periodic garbage collection, to reduce unexpected memory consumption of JVM. Also, we analyze performance effects of the number of cores in a virtual machine (VM) and resource sharing among VMs. Under the virtualized environment in cloud computing infrastructure, one of the critical issues is the management of virtualized computing resources. Accordingly, we present the event processing performance on VMs as a function of virtual CPU assignment types and the number of VMs that share virtual CPUs.

show abstract

Dynamic Switching-Frequency Scaling: Scheduling Overcommitted Domains in Xen VMM

Cited by 11 publications

References 8 publications

An Improved Xen Credit Scheduler for I/O Latency-Sensitive Applications on Multicores

An Improved Xen Credit Scheduler for I/O Latency-Sensitive Applications on Multicores

XCollOpts: A Novel Improvement of Network Virtualizations in Xen for I/O-Latency Sensitive Applications on Multicores

An analysis of performance factors on Esper‐based stream big data processing in a virtualized environment

Contact Info

Product

Resources

About