<div dir="ltr"><div>we apologize if you receive multiple copies of this CfP<br></div><div><br></div><div>=================================================================</div><div><br></div><div>CALL FOR PAPERS </div><div>
<br></div><div>9th Workshop on Virtualization in High-Performance Cloud Computing (VHPC '14)</div><div><br></div><div>held in conjunction with Euro-Par 2014, August 25-29, Porto, Portugal</div><div><br></div><div>=================================================================</div>
<div><br></div><div>Date: August 26, 2014</div><div><br></div><div>Workshop URL: <a href="http://vhpc.org">http://vhpc.org</a></div><div><br></div><div>Paper Submission Deadline: May 30, 2014</div><div><br></div><div><br>
</div><div>CALL FOR PAPERS</div><div><br></div><div>Virtualization technologies constitute a key enabling factor for flexible resource</div><div>management in modern data centers, and particularly in cloud environments.</div>
<div>Cloud providers need to dynamically manage complex infrastructures in a</div><div>seamless fashion for varying workloads and hosted applications, independently of</div><div>the customers deploying software or users submitting highly dynamic and</div>
<div>heterogeneous workloads. Thanks to virtualization, we have the ability to manage</div><div>vast computing and networking resources dynamically and close to the marginal</div><div>cost of providing the services, which is unprecedented in the history of scientific</div>
<div>and commercial computing. </div><div><br></div><div>Various virtualization technologies contribute to the overall picture in different</div><div>ways: machine virtualization, with its capability to enable consolidation of multiple</div>
<div>under-utilized servers with heterogeneous software and operating systems (OSes),</div><div>and its capability to live-migrate a fully operating virtual machine (VM) with a very</div><div>short downtime, enables novel and dynamic ways to manage physical servers;</div>
<div>OS-level virtualization, with its capability to isolate multiple user-space</div><div>environments and to allow for their co-existence within the same OS kernel,</div><div>promises to provide many of the advantages of machine virtualization with high</div>
<div>levels of responsiveness and performance; I/O Virtualization allows physical</div><div>NICs/HBAs to take traffic from multiple VMs; network virtualization, with its</div><div>capability to create logical network overlays that are independent of the</div>
<div>underlying physical topology and IP addressing, provides the fundamental</div><div>ground on top of which evolved network services can be realized with an</div><div>unprecedented level of dynamicity and flexibility; the increasingly adopted</div>
<div>paradigm of Software-Defined Networking (SDN) promises to extend this</div><div>flexibility to the control and data planes of network paths. These technologies</div><div>have to be inter-mixed and integrated in an intelligent way, to support</div>
<div>workloads that are increasingly demanding in terms of absolute performance,</div><div>responsiveness and interactivity, and have to respect well-specified Service-</div><div>Level Agreements (SLAs), as needed for industrial-grade provided services.</div>
<div>Indeed, among emerging and increasingly interesting application domains</div><div>for virtualization, we can find big-data application workloads in cloud</div><div>infrastructures, interactive and real-time multimedia services in the cloud,</div>
<div>including real-time big-data streaming platforms such as used in real-time</div><div>analytics supporting nowadays a plethora of application domains. Distributed</div><div>cloud infrastructures promise to offer unprecedented responsiveness levels for</div>
<div>hosted applications, but that is only possible if the underlying virtualization</div><div>technologies can overcome most of the latency impairments typical of current</div><div>virtualized infrastructures (e.g., far worse tail-latency). What is more, in data</div>
<div>communications Network Function Virtualization (NFV) is becoming a key</div><div>technology enabling a shift from supplying hardware-based network functions,</div><div>to providing them in a software-based and elastic way. In conjunction with</div>
<div>(public and private) cloud technologies, NFV may be used for constructing the</div><div>foundation for cost-effective network functions that can easily and seamlessly</div><div>adapt to demand, still keeping their major carrier-grade characteristics in terms</div>
<div>of QoS and reliability.</div><div><br></div><div>The Workshop on Virtualization in High-Performance Cloud Computing (VHPC)</div><div>aims to bring together researchers and industrial practitioners facing the challenges</div>
<div>posed by virtualization in order to foster discussion, collaboration, mutual exchange</div><div>of knowledge and experience, enabling research to ultimately provide novel</div><div>solutions for virtualized computing systems of tomorrow.</div>
<div><br></div><div>The workshop will be one day in length, composed of 20 min paper presentations,</div><div>each followed by 10 min discussion sections, and lightning talks, limited to 5</div><div>minutes. Presentations may be accompanied by interactive demonstrations.</div>
<div><br></div><div>TOPICS</div><div><br></div><div>Topics of interest include, but are not limited to:</div><div><br></div><div>- Management, deployment and monitoring of virtualized environments</div><div>- Language-process virtual machines</div>
<div>- Performance monitoring for virtualized/cloud workloads</div><div>- Virtual machine monitor platforms</div><div>- Topology management and optimization for distributed virtualized applications</div><div>- Paravirtualized I/O</div>
<div>- Improving I/O and network virtualization including use of RDMA, Infiniband, PCIe</div><div>- Improving performance in VM access to GPUs, GPU clusters, GP-GPUs</div><div>- HPC storage virtualization</div><div>- Virtualized systems for big-data and analytics workloads</div>
<div>- Optimizations and enhancements to OS virtualization support</div><div>- Improving OS-level virtualization and its integration within cloud management</div><div>- Performance modelling for virtualized/cloud applications</div>
<div>- Heterogeneous virtualized environments</div><div>- Network virtualization</div><div>- Software defined networking</div><div>- Network function virtualization</div><div>- Hypervisor and network virtualization QoS and SLAs</div>
<div>- Cloudbursting</div><div>- Evolved European grid architectures including such based on network virtualization</div><div>- Workload characterization for VM-based environments</div><div>- Optimized communication libraries/protocols in the cloud</div>
<div>- System and process/bytecode VM convergence</div><div>- Cloud frameworks and APIs</div><div>- Checkpointing/migration of VM-based large compute jobs</div><div>- Job scheduling/control/policy with VMs </div><div>- Instrumentation interfaces and languages</div>
<div>- VMM performance (auto-)tuning on various load types</div><div>- Cloud reliability, fault-tolerance, and security</div><div>- Research, industrial and educational use cases</div><div>- Virtualization in cloud, cluster and grid environments</div>
<div>- Cross-layer VM optimizations</div><div>- Cloud HPC use cases including optimizations</div><div>- Services in cloud HPC </div><div>- Hypervisor extensions and tools for cluster and grid computing</div><div>- Cluster provisioning in the cloud</div>
<div>- Performance and cost modelling</div><div>- Languages for describing highly-distributed compute jobs</div><div>- VM cloud and cluster distribution algorithms, load balancing</div><div>- Instrumentation interfaces and languages</div>
<div>- Energy-aware virtualization</div><div><br></div><div>Important Dates</div><div><br></div><div>Rolling Paper registration</div><div>May 30, 2014 - Full paper submission</div><div>July 4, 2014 - Acceptance notification</div>
<div>October 3, 2014 - Camera-ready version due</div><div><br></div><div>August 26, 2014 - Workshop Date</div><div><br></div><div><br></div><div>TPC</div><div><br></div><div>CHAIR</div><div><br></div><div>Michael Alexander (chair), TU Wien, Austria</div>
<div>Anastassios Nanos (co-chair), NTUA, Greece</div><div>Tommaso Cucinotta (co-chair), Bell Labs, Dublin, Ireland</div><div><br></div><div>PROGRAM COMMITTEE </div><div>Costas Bekas, IBM</div><div>Jakob Blomer, CERN</div>
<div>Roberto Canonico, University of Napoli Federico II, Italy</div><div>Paolo Costa, MS Research Cambridge, England</div><div>Jorge Ejarque Artigas, Barcelona Supercomputing Center, Spain</div><div>William Gardner, University of Guelph, USA</div>
<div>Balazs Gerofi, University of Tokyo, Japan</div><div>Krishna Kant, Temple University, USA</div><div>Romeo Kinzler, IBM</div><div>Nectarios Koziris, National Technical University of Athens, Greece</div><div>Giuseppe Lettieri, University of Pisa, Italy</div>
<div>Jean-Marc Menaud, Ecole des Mines de Nantes, France</div><div>Christine Morin, INRIA, France</div><div>Dimitrios Nikolopoulos, Queen's University of Belfast, UK</div><div>Herbert Poetzl, VServer, Austria</div><div>
Luigi Rizzo, University of Pisa, Italy</div><div>Josh Simons, VMWare, USA</div><div>Borja Sotomayor, University of Chicago, USA</div><div>Vangelis Tasoulas, Simula Research Lab, Norway</div><div>Yoshio Turner, HP Labs, USA</div>
<div>Kurt Tutschku, Blekinge Institute of Technology, Sweden</div><div>Chao-Tung Yang, Tunghai University, Taiwan</div><div><br></div><div><br></div><div>PAPER SUBMISSION-PUBLICATION</div><div><br></div><div>Papers submitted to the workshop will be reviewed by at least two</div>
<div>members of the program committee and external reviewers. Submissions</div><div>should include abstract, key words, the e-mail address of the</div><div>corresponding author, and must not exceed 10 pages, including tables</div>
<div>and figures at a main font size no smaller than 11 point. Submission</div><div>of a paper should be regarded as a commitment that, should the paper</div><div>be accepted, at least one of the authors will register and attend the</div>
<div>conference to present the work.</div><div><br></div><div>Accepted papers will be published in the Springer LNCS series - the</div><div>format must be according to the Springer LNCS Style. Initial</div><div>submissions are in PDF; authors of accepted papers will be requested</div>
<div>to provide source files.</div><div><br></div><div>Format Guidelines:</div><div><a href="http://www.springer.de/comp/lncs/authors.html">http://www.springer.de/comp/lncs/authors.html</a></div><div><br></div><div>EasyChair Abstract Submission Link:</div>
<div><a href="https://www.easychair.org/conferences/?conf=europar2014ws">https://www.easychair.org/conferences/?conf=europar2014ws</a></div><div><br></div><div>GENERAL INFORMATION</div><div><br></div><div>The workshop is one day in length and will be held in conjunction with</div>
<div>Euro-Par 2014, 25-29 August, Porto, Portugal</div><div><br></div></div>