Det här arbetet hade målet att undersöka vilken storlek förlusten av resurser som Hyper-V och Windows Server container introducerar när ett system delas upp med dessa. Genom att testa detta skulle man också kunna jämföra Microsofts implementering av containerbaserad virtualisering med Linux implementering och se om det finns några likheter. För att kunna jämföra med tidigare forskning fanns målet att försöka efterlikna metoden i tidigare forskning som gjort liknande tester i operativsystemet Linux med KVM och LXC så nära som möjligt. Resultaten visades sig till stor del överensstämma med resultaten på tester genomförda i Linux. Det vill säga att container-baserad virtualisering har överlag en mindre förlust av resurser än hypervisor-baserad virtualisering. Undantaget var dock nätverkstesterna där container-baserad virtualisering presterade sämst men även detta överensstämde med tidigare forskning. / This work had the goal of looking at the loss of resources that Hyper-V and Windows Server container introduce when a system is partitioned with these. By testing this, you could also compare Microsoft's implementation of container-based virtualization with Linux implementation and see if there are any similarities. To be able to compare with previous research, the goal was to try to mimic the method in previous research that made similar tests in the Linux operating system with KVM and LXC as close as possible. The results were largely consistent with the results of tests conducted in Linux. That is, container-based virtualization generally has a smaller loss of resources than hypervisor-based virtualization. The exception was, however, the network tests where container-based virtualization performed poorest, but this also was consistent with previous research.

Several cloud providers offer reltional databases as part of their portfolio. It is however not obvious how resource virtualization and sharing, which is inherent to cloud computing, influence performance and predictability of these cloud databases. Cloud providers give little to no guarantees for consistent execution or isolation from other users. To evaluate the performance isolation capabilities of two commercial cloud databases, we ran a series of experiments over the course of a week (a query, a minute) and report variations in query response times. As a baseline, we ran the same experiments on a dedicated server in our data center. The results show that in the cloud single outliers are up to 31 times slower than the average. Additionally, one can see a point in time after which the average performance of all executed queries improves by 38 %.

ITC/USA 2009 Conference Proceedings / The Forty-Fifth Annual International Telemetering Conference and Technical Exhibition / October 26-29, 2009 / Riviera Hotel & Convention Center, Las Vegas, Nevada / Virtualization technologies have the potential to transform the telemetry post-processing environment. Significant efficiencies can be gained by migrating telemetry post processing activities to virtual computing platforms. However, while facilitating better server utilization, virtualization also presents several challenges; one of the most difficult of those challenges being security. In virtualization, server environments are replicated in software; unfortunately, the security individual servers provide is not replicated in a software stack implementation of a server environment. After reviewing virtualization fundamentals, security issues and their impact on telemetry post processing will be discussed.

The prevalence of multi-core processors with recent advancement in virtualization technologies has enabled horizontal and vertical scaling within a physical node achieving economical sharing of computing infrastructures as computing clouds. Through hardware virtualization, consolidated servers each with specific number of core allotment run on the same physical node in dedicated Virtual Machines (VMs) to increase overall node utilization which increases profit by reducing operational costs. Unfortunately, despite the conceptual simplicity of vertical and horizontal scaling in virtualized cloud environments, leveraging the full potential of this technology has presented significant scalability challenges in practice. One of the fundamental problems is the performance unpredictability in virtualized cloud environments (ranked fifth in the top 10 obstacles for growth of cloud computing). In this dissertation, we present two case studies in vertical and horizontal scaling to this challenging problem. For the first case study, we describe concrete experimental evidence that shows important source of performance variations: mapping of virtual CPU to physical cores. We then conduct an experimental comparative study of three major hypervisors (i.e., VMware, KVM, Xen) with regard to their support of n-tier applications running on multi-core processor. For the second case study, we present empirical study that shows memory thrashing caused by interference among consolidated VMs is a significant source of performance interference that hampers horizontal scalability of an n-tier application performance. We then execute transient event analyses of fine-grained experiment data that link very short bottlenecks with memory thrashing to the very long response time (VLRT) requests. Furthermore we provide three practical techniques such as VM migration, memory reallocation, soft resource allocation and show that they can mitigate the effects of performance interference among consolidate VMs.

<p>This thesis presents an evaluation of the Solaris version of the Xen virtual machine monitor and a comparison of its performance to the performance of Solaris Containers under similar conditions. Xen is a virtual machine monitor, based on the paravirtualization approach, which provides an instruction set different to the native machine environment and therefore requires modifications to the guest operating systems. Solaris Zones is an operating system-level virtualization technology that is part of the Solaris OS. Furthermore, we provide a basic performance evaluation of the security modules for Xen and Zones, known as sHype and Solaris Trusted Extensions, respectively.</p><p>We evaluate the control domain (know as Domain-0) and the user domain performance as the number of user domains increases. Testing Domain-0 with an increasing number of user domains allows us to evaluate how much overhead virtual operating systems impose in the idle state and how their number influences the overall system performance. Testing one user domain and increasing the number of idle domains allows us to evaluate how the number of domains influences operating system performance. Testing concurrently loaded increasing numbers of user domains we investigate total system efficiency and load balancing dependent on the number of running systems.</p><p>System performance was limited by CPU, memory, and hard drive characteristics. In the case of CPU-bound tests Xen exhibited performance close to the performance of Zones and to the native Solaris performance, loosing 2-3% due to the virtualization overhead. In case of memory-bound and hard drive-bound tests Xen showed 5 to 10 times worse performance.</p>

Medan servervirtualisering växt väldigt mycket det senaste decenniet har Apple nyligen börjat tillåta virtualisering av deras operativsystem Mac OS X. På grund av Apples stora framgångar med att lyckas sälja sina produkter till hemanvändare har nu även Mac-datorer börjat hitta till arbetsplatsen. Detta gör att det kommer finnas ett behov av att virtualisera Mac OS X i företagsmiljöer i framtiden då servervirtualisering är väldigt kostnadseffektivt. För att kunna virtualisera krävs en så kallad hypervisor. Det finns två olika typer, typ 1 och typ 2. I denna uppsats ville vi undersöka prestandaskillnader mellan dessa två typer vid virtualisering av Mac OS X. Detta för att kunna se hur prestandan skiljer dem sinsemellan. Prestandatester genomfördes i en experimentmiljö där resultaten visar att en typ 1-hypervisor ger mindre prestandaförluster än en typ 2- hypervisor.

Network virtualization is becoming a fundamental building block of future Internet architectures. By adding networking resources into the “cloud”, it is possible for users to rent virtual routers from the underlying network infrastructure, connect them with virtual channels to form a virtual network, and tailor the virtual network (e.g., load application-specific networking protocols, libraries and software stacks on to the virtual routers) to carry out a specific task. In addition, network virtualization technology allows such special-purpose virtual networks to co-exist on the same set of network infrastructure without interfering with each other. Although the underlying network resources needed to support virtualized networks are rapidly becoming available, constructing a virtual network from the ground up and using the network is a challenging and labor-intensive task, one best left to experts. To tackle this problem, we introduce the concept of a HyperNet, a pre-built, pre-configured network package that a user can easily deploy or access a virtual network to carry out a specific task (e.g., multicast video conferencing). HyperNets package together the network topology configuration, software, and network services needed to create and deploy a custom virtual network. Users download HyperNets from HyperNet repositories and then “run” them on virtualized network infrastructure much like users download and run virtual appliances on a virtual machine. To support the HyperNet abstraction, we created a Network Hypervisor service that provides a set of APIs that can be called to create a virtual network with certain characteristics. To evaluate the HyperNet architecture, we implemented several example Hyper-Nets and ran them on our prototype implementation of the Network Hypervisor. Our experiments show that the Hypervisor API can be used to compose almost any special-purpose network – networks capable of carrying out functions that the current Internet does not provide. Moreover, the design of our HyperNet architecture is highly extensible, enabling developers to write high-level libraries (using the Network Hypervisor APIs) to achieve complicated tasks.

Today large software systems (i.e., giants) thrive in commodity markets, but are untrustworthy due to their numerous and inevitable software bugs that can be exploited by the adversary. Thus, the best hope of security is that some small, simple, and trustworthy software components (i.e., wimps) can be protected from attacks launched by adversary-controlled giants. However, wimps in isolation typically give up a variety of basic services (e.g., file system, networking, device I/O), trading usefulness and viability with security. Among these basic services, isolated I/O channels remained an unmet challenge over the past three decades. Isolated I/O is a critical security primitive for a myriad of applications (e.g., secure user interface, remote device control). With this primitive, isolated wimps can transfer I/O data to commodity peripheral devices and the data secrecy and authenticity are protected from the co-existing giants. This thesis addresses this challenge by proposing a new security architecture to provide services to isolated wimps. Instead of restructuring the giants or bloating the Trusted Computing Base that underpins wimp-giant isolation (dubbed underlying TCB), this thesis presents a unique on-demand I/O isolation model and a trusted add-on component called wimpy kernel to instantiate this model. In our model, the wimpy kernel dynamically takes control of the devices managed by a commodity OS, connects them to the isolated wimps, and relinquishes controls to the OS when done. This model creates ample opportunities for the wimpy kernel to outsource I/O subsystem functions to the untrusted OS and verify their results. The wimpy kernel further exports device drivers and I/O subsystem code to wimps and mediates the operations of the exported code. These two methodologies help to significantly reduce the size and complexity of the wimpy kernel for high security assurance. Using the popular and complex USB subsystem as a case study, this thesis illustrates the dramatic reduction of the wimpy kernel; i.e., over 99% of the Linux USB code base is removed. In addition, the wimpy kernel also composes with the underlying TCB, by retaining its code size, complexity and security properties.

Nowadays, Industrial Control Systems (ICSs) are becoming larger and implement more complex functions. Therefore, technologies that are currently used to implement these functions, like hardware platforms and communication protocols might soon become unusable due to the lack of resources. The industry is trying to adopt new technologies that will allow these functionalities to be developed without an increase in the size of the equipment, or of the development costs. To enumerate some of these technologies: virtualization, multi-core technologies are the ones that show the biggest potential. Because these technologies are not mature, research has to be done in order to fully maximize their potential. Another technology that is highly used by the industry is the Ethernet communication protocol. It presents some advantages, but due to the non-real-time nature of the applications that it was designed for, it has to be extended in order to be used in real-time applications. The objective of this thesis work is to model an Ethernet network comprised of software partitions so that it can provide timing guarantees for the traffic that traverses the network. A Response Time Analysis for real-time flows over such networks is proposed. The model and the RTA are evaluated by experiments.

Face au coût de la gestion locale des infrastructures informatiques, de nombreuses entreprises ont décidé de la faire gérer par des fournisseurs externes. Ces derniers, connus sous le nom de IaaS (Infrastructure as a Service), mettent des ressources à la disposition des entreprises sous forme de machine virtuelle (VM - Virtual Machine). Ainsi, les entreprises n'utilisent qu'un nombre limité de machines virtuelles capables de satisfaire leur besoin. Ce qui contribue à la réduction des coûts de l'infrastructure informatique des entreprises clientes. Cependant, cette externalisation soulève pour le fournisseur, les problèmes de respect d'accord de niveau de service (SLA - Service Layer Agreement) souscrit par le client et d'optimisation de la consommation énergétique de son infrastructure. Au regard de l'importance que revêt ces deux défis, de nombreux travaux de recherches se sont intéressés à cette problématique. Les solutions de gestion d'énergie proposées consistent à faire varier la vitesse d'exécution des périphériques concernés. Cette variation de vitesse est implémentée, soit de façon native parce que le périphérique dispose des mécaniques intégrés, soit par simulation à travers des regroupements (spatial et temporel) des traitements. Toutefois, cette variation de vitesse permet d'optimiser la consommation énergétique d'un périphérique mais, a pour effet de bord d'impacter le niveau de service des clients. Cette situation entraine une incompatibilité entre les politiques de variation de vitesse pour la baisse d'énergie et le respect de l'accord de niveau de service. Dans cette thèse, nous étudions la conception et l'implantation d'un gestionnaire de ressources "éco énergétique" dans un système virtualisé. Un tel gestionnaire doit permettre un partage équitable des ressources entre les machines virtuelles tout en assurant une utilisation optimale de l'énergie que consomment ces ressources. Nous illustrons notre étude avec un ordonnanceur de machines virtuelles. La politique de variation de vitesse est implantée par le DVFS (Dynamic Voltage Frequency Scaling) et l'allocation de la capacité CPU aux machines virtuelles l'accord de niveau de service à respecter. / Considering the cost of local management of the computing infrastructures, numerous companies decided to delegate theirs to providers. These latter are known as an Infrastructure as a Service (IaaS) and provide resources to companies in the form of virtual machine (VM). This decision to outsource contributes to lower the cost of IT infrastructure of the customer companies. However, it raises for the provider, the problems of the respect of the Service Layer agreement (SLA) of the customer and of the optimization of the energy consumption of his infrastructure. With regard to the importance of these two challenges, many research works have focused on this problem. The proposed energy management solutions consist in varying the execution speed of the affected devices. This variation of speed is implemented either natively because the device has integrated mechanics, or by simulation through a spatial or temporal batching requests. However, this variation of speed optimizes the energy consumption of a device but has the side effect of degrading the customers SLA. In this thesis, we study the design and the implementation of an energy-efficient resources manager in a virtualized system. Such a manager must ensure a fair share of resources among VMs while ensuring optimal use of the energy consumed by the resources. We illustrate our study thanks to a scheduler of VMs. The DVFS constitutes our energy management policy and the CPU capacity of the VMs the SLA to respect.