KVM, Xen and Docker: A performance analysis for ARM based NFV and cloud computing

Raho, Moritz; Spyridakis, Alexander; Paolino, Michele; Raho, Daniel

doi:10.1109/aieee.2015.7367280

Cited by 56 publications

(32 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The performance of KVM, Xen (VM-based) and Docker on ARM has been compared with the results obtained in Ref. [27]. The result shows that Docker still has the advantage except in file-system operations.…”

Section: Virtualization Based Isolation On Embedded Systemmentioning

confidence: 94%

A Light-weight Multilevel Recoverable Container for Event-driven System: A Self-healing CPS Approach

Zhou¹,

Zuo²,

Hou³

et al. 2017

Proceedings of the 3rd International Conference on Wireless Communication and Sensor Networks (WCSN 2016)

View full text Add to dashboard Cite

Abstract-Cyber Physical Systems (CPS) is regarded as a new technological revolution, which tightly integrates computing, communication, and control technologies, to build a kind of smart networked distributed embedded control system. CPS is designed to interact autonomously with the volatile external environment, which implies that the requirements is constantly changing during run-time. So guaranteeing the reliability of system becomes extremely difficult. Flexible self-healing mechanisms are needed urgently to improve the reliability and availability of CPS. This paper presents a light-weight container-based virtualization for event-driven CPS. By providing a unique run-time stack for each application, the container isolates faults and limits the effect of failures. Furthermore, a multilevel fault detection and recovery method is integrated to protect applications and to limit the fault propagation. And the analysis shows the container has very low memory footprint (939 bytes) and constant performance overhead. Also the testing manifests that the multilevel recovery is high reliable on WCET violation failure recovery even if the application is not well designed or malicious.

show abstract

Section: Virtualization Based Isolation On Embedded Systemmentioning

confidence: 94%

A Light-weight Multilevel Recoverable Container for Event-driven System: A Self-healing CPS Approach

Zhou¹,

Zuo²,

Hou³

et al. 2017

Proceedings of the 3rd International Conference on Wireless Communication and Sensor Networks (WCSN 2016)

View full text Add to dashboard Cite

show abstract

“…Recent studies demonstrate that the performance of VMs is approaching near-native performance [12]. This trend of virtualization, coupled with its lower cost when compared to investing in physical replicas, has made it an obvious choice for redundancy in fault-tolerance architectures [3,5,7].…”

Section: Related Workmentioning

confidence: 99%

“…Physical controller replication increases the cost of deployment and maintenance. Advances in virtualization technology enable virtual machines (VMs) to provide near-native performance [12]. Such performance, coupled with the low cost of VMs, has led to the use of virtualization-based replication schemes, whereby multiple virtual-machine replicas (hereafter referred to as virtual replicas) are deployed on the same physical node [3,5,7].…”

Section: Introductionmentioning

confidence: 99%

Experimental validation of the suitability of virtualization-based replication for fault tolerance in real-time control of electric grids

Kohroudi

Mostafa

Mohiuddin

et al. 2018

Proceedings of the 12th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement

View full text Add to dashboard Cite

Real-time control systems (RTCSs) perform complex control and require low response times. They typically use third-party software libraries and are deployed on generic hardware, which suffer from delay faults that can cause serious damage. To improve availability and latency, the controllers in RTCSs are replicated on physical nodes. As physical replication is expensive, we study the alternative of exploiting virtualization technology to run multiple virtual replicas on the same physical node. As virtual replicas share the same resources, the delay faults they experience might be correlated, which would make such a replication method unsuitable. We conduct several experiments with an RTCS for electric grids, with multiple virtual replicas of its controller. We find that although the delay of a virtual machine is higher than of a physical machine, the correlation between high delays among the virtual replicas is insignificant, causing an overall improved availability. We conclude that virtual replication is indeed applicable to certain RTCSs, as it can improve reliability without added cost. CCS CONCEPTS • Computer systems organization → Availability; • Software and its engineering → Real-time systems software;

show abstract

“…Spoiala et al built the WebRTC over KVM and Docker [10]. Virtualization about cloud computing is evaluated in [29][30][31]. These articles showed that Docker outperforms KVM in most experiments because Docker is lite.…”

Section: Related Workmentioning

confidence: 99%

Empirical Experience and Experimental Evaluation of Open5GCore over Hypervisor and Container

Chang

Qiu

Chen

et al. 2018

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

One of the most important technologies for future 5G networks is to utilize Network Function Virtualization (NFV) to virtualize the network components. NFV provides flexibility, short time to market, and low cost solution to build network services, which are important features of 5G networks. Although the idea of virtualization is just being applied to cellular networks, it has been used in the community of cloud computing. There are two main virtualization techniques, hypervisor and container. In this paper, we present our practical experience of virtualizing Open5GCore, a commercial product of SDN-enabled Evolved Packet Core (EPC), over hypervisor and container. In addition to describing how to virtualize Open5GCore, we also present the experimental performance evaluation of the systems. Finally, some important lessons learned are provided.

show abstract

KVM, Xen and Docker: A performance analysis for ARM based NFV and cloud computing

Cited by 56 publications

References 21 publications

A Light-weight Multilevel Recoverable Container for Event-driven System: A Self-healing CPS Approach

A Light-weight Multilevel Recoverable Container for Event-driven System: A Self-healing CPS Approach

Experimental validation of the suitability of virtualization-based replication for fault tolerance in real-time control of electric grids

Empirical Experience and Experimental Evaluation of Open5GCore over Hypervisor and Container

Contact Info

Product

Resources

About