Cloud computing makes extensive use of virtual machines because they permit workloads to be isolated from one another and for the resource usage to be somewhat controlled. In this paper, we explore the performance of traditional virtual machine (VM) deployments, and contrast them with the use of Linux containers. We use KVM as a representative hypervisor and Docker as a container manager. Our results show that containers result in equal or better performance than VMs in almost all cases. Both VMs and containers require tuning to support I/Ointensive applications. We also discuss the implications of our performance results for future cloud architectures.

/pdf/an-updated-performance-comparison-of-virtual-machines-and-51dkcxz9au.pdf

An updated performance comparison of virtual machines and Linux containers

As Cloud computing emerges as a dominant paradigm in distributed systems, it is important to fully understand the underlying technologies that make Clouds possible. One technology, and perhaps the most important, is virtualization. Recently virtualization, through the use of hyper visors, has become widely used and well understood by many. However, there are a large spread of different hyper visors, each with their own advantages and disadvantages. This paper provides an in-depth analysis of some of today's commonly accepted virtualization technologies from feature comparison to performance analysis, focusing on the applicability to High Performance Computing environments using Future Grid resources. The results indicate virtualization sometimes introduces slight performance impacts depending on the hyper visor type, however the benefits of such technologies are profound and not all virtualization technologies are equal. From our experience, the KVM hyper visor is the optimal choice for supporting HPC applications within a Cloud infrastructure.

/pdf/analysis-of-virtualization-technologies-for-high-performance-1obldolfj7.pdf

Analysis of Virtualization Technologies for High Performance Computing Environments

Intrusion detection techniques in cloud environment

Background— The standard interpretation of perfusion imaging is based on the assessment of relative perfusion distribution. The limitations of that approach have been recognized in patients with multivessel disease and endothelial dysfunction. To date, however, no large clinical studies have investigated the value of measuring quantitative blood flow and compared that with relative uptake.

Methods and Results— One hundred four patients with moderate (30%–70%) pretest likelihood of coronary artery disease (CAD) underwent PET imaging during adenosine stress using 15O-water and dynamic imaging. Absolute myocardial blood flow was calculated from which both standard relative myocardial perfusion images and images scaled to a known absolute scale were produced. The patients and the regions then were classified as normal or abnormal and compared against the reference of conventional angiography with fractional flow reserve. In patient-based analysis, the positive predictive value, negative predictive value, and accuracy of absolute perfusion in the detection of any obstructive CAD were 86%, 97%, and 92%, respectively, with absolute quantification. The corresponding values with relative analysis were 61%, 83%, and 73%, respectively. In region-based analysis, the receiver operating characteristic curves confirmed that the absolute quantification was superior to relative assessment. In particular, the specificity and positive predictive value were low using just relative differences in flow. Only 9 of 24 patients with 3-vessel disease were correctly assessed using relative analysis.

Conclusions— The measurement of myocardial blood flow in absolute terms has a significant impact on the interpretation of myocardial perfusion. As expected, multivessel disease is more accurately detected.

Clinical Trial Registration— URL: . Unique identifier: [NCT00627172][1].

 [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT00627172&atom=%2Fcirccvim%2F4%2F6%2F678.atom

Clinical value of absolute quantification of myocardial perfusion with (15)O-water in coronary artery disease.

The networking equipment market is being transformed by the need for greater openness and flexibility, not only for research purposes but also for in-house innovation by the equipment owners. In contrast to networking gear following the model of computer mainframes, where closed software runs on proprietary hardware, the software-defined networking approach effectively decouples the data from the control plane via an open API (i.e., OpenFlow protocol) that allows the (remote) control of packet forwarding engines. Motivated by this scenario, we propose RouteFlow, a commodity routing architecture that combines the line-rate performance of commercial hardware with the flexibility of open-source routing stacks (remotely) running on general purpose computers. The outcome is a novel point in the design space of commodity routing solutions with far-reaching implications towards virtual routers and IP networks as a service. This paper documents the progress achieved in the design and prototype implementation of our work and outlines our research agenda that calls for a community-driven approach.

http://ecee.colorado.edu/~ekeller/classes/fall2012_sdn/papers/routeflow_cfi_2011_nascimento.pdf

Virtual routers as a service: the RouteFlow approach leveraging software-defined networks

The success of server virtualization has let to the deployment of a huge number of virtual machines in today's data centers, making a manual virtualization management very labor-intensive. The development of appropriate management solutions is hindered by the various management interfaces of different hypervisors. Therefore, a uniform management can be simplified by a layer abstracting from these dedicated hypervisor interfaces. The libvirt management library provides such an interface to different hypervisors. Unfortunately, remote hypervisor management using libvirt has not been possible without altering the managed servers. To overcome this limitation, we have integrated remote hypervisor management facilities into the libvirt driver infrastructure for VMware ESX and Microsoft Hyper-V. This paper presents the resulting architecture as well as experiences gained during the implementation process.

Non-intrusive virtualization management using libvirt

Aim Dynamic perfusion PET offers a clinical relevant advantage over myocardial perfusion scintigraphy due to its ability to measure myocardial blood flow quantitatively This leads to an improved detection of multivessel disease and the possibility to assess not only the culprit lesion but lower grade stenoses as well For appropriate revascularization, perfusion defects must be matched to coronary lesions It has been shown that image fusion of morphological and functional images is superior to side-by-side analysis Still, software for quantitative perfusion PET combined with CT angiography is rare In this paper we present a new software tool for image fusion and visualization of quantitative perfusion PET and coronary morphology derived from CT angiography Methods In our software, a PET uptake image is used for manual co-registration Co-registration results are then applied to the functional data derived from compartment modelling To evaluate the reproducibility of the manual co-registration, we calculated the deviation between a series of manual co-registrations performed on nine pairs of unregistered PET and CT datasets by five trained participants Two dimensional transfer functions were used to highlight the coronary arteries from the CT study in the combined data sets Results The average Euclidian distances for three references points were between 37 and 41 mm The maximum distance was 106 mm By the use of the two dimensional transfer functions, coronary anatomy could be easily visualised either by user-interaction or automatically by use of neuronal networks Conclusions With this approach it is possible to combine quantitative perfusion PET with coronary anatomy derived from CT angiography Our first experiences indicate that manual image fusion with our tool is reproducible and that visualisation of the combined datasets is achieved within short time

Quantitative myocardial perfusion PET combined with coronary anatomy derived from CT angiography: validation of a new fusion and visualisation software.

In the last three years a number of multi-volume GPU ray casting systems have been presented. Some of them are very powerful and provide a wide variety of features. However, these approaches are either only capable of displaying multiple modalities together without logically combining them or they lack the necessary flexibility for rapid visual development. These features are fundamental for visualizing the coherent information of multimodal data. In this paper we therefore present an integrated way of visually specifying a volume rendering pipeline including a flexible multimodal compositing of sampling, transfer functions, logical operators and shading. As a result the data flow can be visually constructed and retraced from preprocessing through to the shader operations. Hence intuitive visual prototyping of multimodal transfer function compositing is possible at runtime.

Seamless Integration of Multimodal Shader Compositing into a Flexible Ray Casting Pipeline

/pdf/effect-of-symmetry-reduction-on-the-magnetic-properties-of-2sq578kp.pdf

Effect of symmetry reduction on the magnetic properties of LnIr<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" id="d1e2666" altimg="si69.svg"><mml:msub><mml:mrow /><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub></mml:math>Si<mml:math xmlns:mml="http://www.w3.org/1998/Mat

Dissections in anatomic heart illustrations are an effective visualization technique and support a variety of applications in exploration and communication. In this paper we implemented a set of volume deformations imitating dissections found in anatomy atlases of the human heart based on individual patient data. This allows physicians and surgeons to compare, explore, and discuss volume data in view of atlas illustrations. The main challenge for these illustrations is the flexible real-time combination of various (geometric) deformations and shadows (influenced by the deformations) to depict shapes and structures. The proposed technical solution to these volume visualizations is a novel GPU-based processing technique for dataflow-based shading graphs. Hence, any deformations can easily be combined in our system with immediate influence on shadows and shading. We show the effectiveness and applicability of our approach by imitating illustrations of heart dissections taken from an anatomy atlas using the patient’s individual volume data.

Matthias Bolte

Papers

Non-intrusive virtualization management using libvirt

Quantitative myocardial perfusion PET combined with coronary anatomy derived from CT angiography: validation of a new fusion and visualisation software.

Seamless Integration of Multimodal Shader Compositing into a Flexible Ray Casting Pipeline

Effect of symmetry reduction on the magnetic properties of LnIr<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" id="d1e2666" altimg="si69.svg"><mml:msub><mml:mrow /><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub></mml:math>Si<mml:math xmlns:mml="http://www.w3.org/1998/Mat

Visualizing Dissections of the Heart in a Dataflow-based Shader Framework for Volume Rendering