Hypervisors vs. Lightweight Virtualization: A Performance Comparison

Morabito, Roberto; Kjallman, Jimmy; Komu, Miika

doi:10.1109/ic2e.2015.74

Cited by 298 publications

(169 citation statements)

References 10 publications

Supporting

Mentioning

146

Contrasting

Unclassified

Order By: Relevance

“…The comparison is based on the performance metrics collected by the benchmarking tools. A similar study, aimed at comparing the performance of containers with hypervisors is [13]. The authors use a set of benchmark, and only the metrics evaluated by the benchmark tools, to assess the performance of Docker, KVM and LXC.…”

Section: Related Workmentioning

confidence: 99%

Measuring Docker Performance

Casalicchio

Perciballi

2017

Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering Companion

View full text Add to dashboard Cite

Today, a new technology is going to change the way platforms for the internet of services are designed and managed. This technology is called container (e.g. Docker and LXC). The internet of service industry is adopting the container technology both for internal usage and as commercial offering. The use of container as base technology for largescale systems opens many challenges in the area of resource management at run-time, for example: autoscaling, optimal deployment and monitoring. Specifically, monitoring of container based systems is at the ground of any resource management solution, and it is the focus of this work. This paper explores the tools available to measure the performance of Docker from the perspective of the host operating system and of the virtualization environment, and it provides a characterization of the CPU and disk I/O overhead introduced by containers.

show abstract

Section: Related Workmentioning

confidence: 99%

Measuring Docker Performance

Casalicchio

Perciballi

2017

Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering Companion

View full text Add to dashboard Cite

show abstract

“…Existing empirical studies related specifically to Docker typically focus on performance aspects, often comparing container performance and overhead with traditional virtualization techniques [22], [23], [24]. However, despite this shortage of empirical work, the importance of Docker for academia and industry is rarely doubted in literature.…”

Section: Related Workmentioning

confidence: 99%

An empirical analysis of the Docker container ecosystem on GitHub

Cito

Schermann

Wittern

et al. 2017

Preprint

View full text Add to dashboard Cite

Abstract-Docker allows packaging an application with its dependencies into a standardized, self-contained unit (a so-called container), which can be used for software development and to run the application on any system. Dockerfiles are declarative definitions of an environment that aim to enable reproducible builds of the container. They can often be found in source code repositories and enable the hosted software to come to life in its execution environment. We conduct an exploratory empirical study with the goal of characterizing the Docker ecosystem, prevalent quality issues, and the evolution of Dockerfiles. We base our study on a data set of over 70000 Dockerfiles, and contrast this general population with samplings that contain the Top-100 and Top-1000 most popular Docker-using projects. We find that most quality issues (28.6%) arise from missing version pinning (i.e., specifying a concrete version for dependencies). Further, we were not able to build 34% of Dockerfiles from a representative sample of 560 projects. Integrating quality checks, e.g., to issue version pinning warnings, into the container build process could result into more reproducible builds. The most popular projects change more often than the rest of the Docker population, with 5.81 revisions per year and 5 lines of code changed on average. Most changes deal with dependencies, that are currently stored in a rather unstructured manner. We propose to introduce an abstraction that, for instance, could deal with the intricacies of different package managers and could improve migration to more light-weight images.

show abstract

“…To face this issue, container‐based virtualization is considered one of the most promising solutions for MEC environments . Containers allow for avoiding the overhead due to virtualized hardware requested by hypervisor‐based virtualization, thus enabling fast initialization and dense deployment of services, as experimented in a previous study . All these features make containers extremely attractive for usage not only in datacentres, but also in edge nodes, such as cloudlets and IoT gateways .…”

Section: Introductionmentioning

confidence: 97%

Providing ultra‐short latency to user‐centric 5G applications at the mobile network edge

Farris

Taleb

Flinck

2017

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Mobile Edge Computing (MEC) will play a key role in next-generation mobile networks to extend the range of supported delay-sensitive applications. Furthermore, an increasing attention is paid to provide user-centric services, to better address the strict requirements of novel immersive applications. In this scenario, MEC solutions need to efficiently cope with user mobility, which requires fast relocation of service instances to guarantee the desired Quality of Experience. However, service migration is still an open issue, especially for resource-constrained edge nodes interconnected by high-latency and low-bandwidth links. In this paper, by leveraging the potential of lightweight container-based virtualization techniques, we investigate a novel approach to support service provisioning in dynamic MEC environments. In particular, we present a framework where proactive service replication for stateless applications is exploited to drastically reduce the time of service migration between different cloudlets and to meet the latency requirements. The performance evaluation shows promising results of our approach with respect to classic reactive service migration. INTRODUCTIONSeveral classes of new applications are challenging the current network and cloud infrastructures. As illustrated in a recent report from International Telecommunication Union, 1 applications such as tactile internet, mobile gaming, and augmented reality are pushing new requirements, especially in terms of latency. To guarantee the desired Quality of Experience (QoE) for the end user, not only the network access delay, but also the cloud service processing should be performed within a few milliseconds. To achieve the 1 ms latency dream for the next-generation 5G cloud-enabled services, both network and cloud providers should jointly cooperate and investigate novel approaches. In this regard, academic and industrial research communities have focused on Mobile Edge Computing (MEC) solutions to exploit processing and storage capabilities at the edge of the network, as near as possible to the end user. 2,3 Indeed, the deployment of micro datacentres in the network access points, known as cloudlets, 4,5 can guarantee remarkable benefits of low latency interaction and scalability, by balancing the workload over the distributed edge infrastructure. Furthermore, bringing the cloudlet concept into the Internet of Things (IoT) scenario results in the definition of the so-called Fog Computing paradigm, which envisages a highly virtualized infrastructure composed of distributed edge nodes to monitor and analyse the most time-sensitive data generated by network connected devices. [6][7][8] In a MEC environment, a key challenge is represented by the user mobility, which could cause significant application degradation, even in small scenarios, as reported in a previous study.9,10 To guarantee service continuity and to meet the strict requirements of application latency, the MEC framework needs to properly cope with service migration between edge nodes. I...

show abstract

Hypervisors vs. Lightweight Virtualization: A Performance Comparison

Cited by 298 publications

References 10 publications

Measuring Docker Performance

Measuring Docker Performance

An empirical analysis of the Docker container ecosystem on GitHub

Providing ultra‐short latency to user‐centric 5G applications at the mobile network edge

Contact Info

Product

Resources

About