Virtualization Technologies in Distributed Computing最新文献

Evaluating hardware-assisted virtualization for deploying HPC-as-a-service 评估部署hpc即服务的硬件辅助虚拟化

Virtualization Technologies in Distributed Computing Pub Date : 2013-06-18 DOI: 10.1145/2465829.2465833

H. Palit, Xiaorong Li, Sifei Lu, L. C. Larsen, Joseph A. Setia

{"title":"Evaluating hardware-assisted virtualization for deploying HPC-as-a-service","authors":"H. Palit, Xiaorong Li, Sifei Lu, L. C. Larsen, Joseph A. Setia","doi":"10.1145/2465829.2465833","DOIUrl":"https://doi.org/10.1145/2465829.2465833","url":null,"abstract":"Virtualization has been the main driver behind the rise of Cloud computing. Despite Cloud computing's tremendous benefits to many applications (e.g., enterprise, Web, game/ multimedia, life sciences, and data analytics), its success in High Performance Computing (HPC) domain has been limited. The oft-cited reason is, apparently, latency caused by virtualization. Meanwhile, the rising popularity of virtualization has compelled CPU vendors to incorporate virtualization technology (VT) in chips. This hardware VT is believed to accelerate context switching, speed up memory address translation, and enable I/O direct access; those are basically sources of virtualization overheads. This paper reports the evaluation on computation and communication performance of different virtualized environments, i.e., Xen and KVM, leveraging hardware VT. Different network fabrics, namely Gigabit Ethernet and InfiniBand, were employed and tested in the virtualized environments and their results were compared against those in the native environments. A real-world HPC application (an MPI-based hydrodynamic simulation) was also used to assess the performance. Outcomes indicate that hardware-assisted virtualization can bring HPC-as-a-Service into realization.","PeriodicalId":176127,"journal":{"name":"Virtualization Technologies in Distributed Computing","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126604953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 12

Bridging the tenant-provider gap in networked cloud services 弥合网络云服务中的租户-提供商差距

Virtualization Technologies in Distributed Computing Pub Date : 2013-06-18 DOI: 10.1145/2465829.2465836

Paolo Costa

{"title":"Bridging the tenant-provider gap in networked cloud services","authors":"Paolo Costa","doi":"10.1145/2465829.2465836","DOIUrl":"https://doi.org/10.1145/2465829.2465836","url":null,"abstract":"The shared nature of the network in today's multi-tenant data centers implies that network performance for tenants can vary significantly. Network performance variability hurts application performance, which makes tenant costs unpredictable and causes provider revenue loss. To address these issues, we proposed to extend the interface tenant-provider to include tenant's network requirements. We designed two virtual network abstractions that capture the trade-off between tenant's expressiveness and provider's flexibility, providing performance isolation and predictability. Further, we investigated how to allow tenants to express high-level goals and automatically predict the resources needed to achieve them. Since multiple resource combinations may achieve the same goal, our system chooses the combination most suitable for the provider. This enables a departure from today's resource-based pricing to a novel job-based pricing model. In this talk, Dr. Paolo Costa will motivate this approach and illustrate the design of the associated system. he will also briefly discuss evaluation results, which show that these new abstractions, by bridging the tenant-provider gap, can enable a symbiotic tenant-provider relationship with significant benefits for both entities.","PeriodicalId":176127,"journal":{"name":"Virtualization Technologies in Distributed Computing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129618262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

NRS: a system for automated network virtualization in iaas cloud infrastructures NRS: iaas云基础设施中的自动化网络虚拟化系统

Virtualization Technologies in Distributed Computing Pub Date : 2013-06-18 DOI: 10.1145/2465829.2465832

D. Theodorou, R. H. Mak, J. J. Keijser, Tristan Suerink

引用次数: 2

Applications-aware virtual machine provisioning 支持应用程序的虚拟机供应

Virtualization Technologies in Distributed Computing Pub Date : 2013-06-18 DOI: 10.1145/2465829.2465837

A. Chandra

{"title":"Applications-aware virtual machine provisioning","authors":"A. Chandra","doi":"10.1145/2465829.2465837","DOIUrl":"https://doi.org/10.1145/2465829.2465837","url":null,"abstract":"Virtualization is widely used in large-scale computing environments, such as clouds, data centers, and grids, to provide multi-tenancy while retaining application isolation and portability. The consumers and providers of such virtualized infrastructures typically have distinct, often conflicting, requirements and optimization metrics. For instance, the user of a virtualized infrastructure typically cares about achieving high performance for their applications while minimizing costs. From the infrastructure provider's perspective, the desired objective is to achieve high system throughput and resource utilization, while minimizing energy consumption. We argue that providers can significantly lower operational costs and improve hosted application performance through application-aware virtual machine provisioning: using application characteristics, such as their data dependencies and resource usage patterns, in VM placement, migration, and resource management decisions. In this talk, Prof. Chandra will illustrate the notion of application-aware virtual machine provisioning through two systems that have been developed as part of his research. Starling is an affinity-aware VM provisioning framework that incorporates intra-application communication patterns to allocate virtual machines on the available physical resources. It dynamically adjusts VM placement to achieve both faster application runtimes as well as lower network communication overheads. STEAMEngine is a provisioning framework for virtualized MapReduce clusters that incorporates MapReduce job characteristics to optimize both cloud-level (e.g., energy) and user-level (e.g., performance) metrics. STEAMEngine exploits spatio-temporal tradeoffs in VM provisioning: efficient spatial fitting of VMs on servers based on their resource requirements, as well as balanced temporal fitting based on the estimated job runtimes","PeriodicalId":176127,"journal":{"name":"Virtualization Technologies in Distributed Computing","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115818910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Streaming as a hypervisor service 流作为一个管理程序服务

Virtualization Technologies in Distributed Computing Pub Date : 2013-06-18 DOI: 10.1145/2465829.2465831

A. Nordal, Åge Kvalnes, Robert Pettersen, D. Johansen

引用次数: 7

New frontiers in cloud computing research 云计算研究的新领域

Virtualization Technologies in Distributed Computing Pub Date : 2013-06-18 DOI: 10.1145/2465829.2465835

H. Lei

{"title":"New frontiers in cloud computing research","authors":"H. Lei","doi":"10.1145/2465829.2465835","DOIUrl":"https://doi.org/10.1145/2465829.2465835","url":null,"abstract":"Cloud computing is challenging conventional wisdom and practices in the management of IT resources and capabilities; at the same time, the technical underpinnings of cloud computing offer unique opportunities and levers for addressing these challenges. In this talk, I will talk about a number of new research thrusts brought forth by the cloud computing paradigm. First, the large magnitude of a typical virtual machine (VM) image and the large volume of common VM libraries have major performance implications for many cloud operations. Fortunately, content similarity that often exists across distinct VM images may be exploited to optimize the transfer and storage of VM images. Second, resource virtualization in the cloud is at odds with the assumptions in traditional software license models, which must be adapted or extended for virtualized environments. There is also a critical need for new license management techniques to help cloud-based software users achieve cost-effectiveness and compliance. Third, cloud infrastructure is evolving towards heterogeneity and is becoming programmable and composable. The so-called Software-Defined Infrastructure promises agility, flexibility, efficiency, and resilience through continuous and proactive optimization of heterogeneous resources to meet the changing requirements of a wide variety of workloads. For each of these research thrusts, I will discuss the specific technical issues and some initial work conducted at IBM Research.","PeriodicalId":176127,"journal":{"name":"Virtualization Technologies in Distributed Computing","volume":"519 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116255919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

VirtualWire: system support for live migrating virtual networks across clouds VirtualWire:系统支持跨云实时迁移虚拟网络

Virtualization Technologies in Distributed Computing Pub Date : 2013-06-18 DOI: 10.1145/2465829.2465838

H. Jamjoom

引用次数: 5

Multi-tenancy on GPGPU-based servers 基于gpgpu的服务器上的多租户

Virtualization Technologies in Distributed Computing Pub Date : 2013-06-18 DOI: 10.1145/2465829.2465830

D. Sengupta, Raghavendra Belapure, K. Schwan

{"title":"Multi-tenancy on GPGPU-based servers","authors":"D. Sengupta, Raghavendra Belapure, K. Schwan","doi":"10.1145/2465829.2465830","DOIUrl":"https://doi.org/10.1145/2465829.2465830","url":null,"abstract":"While GPUs have become prominent both in high performance computing and in online or cloud services, they still appear as explicitly selected 'devices' rather than as first class schedulable entities that can be efficiently shared by diverse server applications. To combat the consequent likely under-utilization of GPUs when used in modern server or cloud settings, we propose 'Rain', a system level abstraction for GPU \"hyperthreading\" that makes it possible to efficiently utilize GPUs without compromising fairness among multiple tenant applications. Rain uses a multi-level GPU scheduler that decomposes the scheduling problem into a combination of load balancing and per-device scheduling. Implemented by overriding applications' standard GPU selection calls, Rain operates without the need for application modification, making possible GPU scheduling methods that include prioritizing certain jobs, guaranteeing fair shares of GPU resources, and/or favoring jobs with least attained GPU services. GPU multi-tenancy via Rain is evaluated with server workloads using a wide variety of CUDA SDK and Rodinia suite benchmarks, on a multi-GPU, multi-core machine typifying future high end server machines. Averaged over ten applications, GPU multi-tenancy on a smaller scale server platform results in application speedups of up to 1.73x compared to their traditional implementation with NVIDIA's CUDA runtime. Averaged over 25 pairs of short and long running applications, on an emulated larger scale server machine, multi-tenancy results in system throughput improvements of up to 6.71x, and in 43% and 29.3% improvements in fairness compared to using the CUDA runtime and a naïve fair-share scheduler.","PeriodicalId":176127,"journal":{"name":"Virtualization Technologies in Distributed Computing","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130404903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 30

Tradeoffs in compressing virtual machine checkpoints 压缩虚拟机检查点的权衡

Virtualization Technologies in Distributed Computing Pub Date : 2013-06-18 DOI: 10.1145/2465829.2465834

Kai-yuan Hou, K. Shin, Yoshio Turner, S. Singhal

引用次数: 6

Elastic IP and security groups implementation using OpenFlow 使用OpenFlow实现弹性IP和安全组

Virtualization Technologies in Distributed Computing Pub Date : 2012-06-18 DOI: 10.1145/2287056.2287069

G. Stabler, Aaron Rosen, S. Goasguen, Kuang-Ching Wang

引用次数: 32