Static provisioning of storage resources may lead to over-provisioning of resources, which increases costs, or under-provisioning, which runs the risk of violating application-level QoS goals. Toward this end, virtualization technologies have made automated provisioning of storage resources easier allowing more effective management of the resources. In this work, we present an approach that suggests a series of dynamic provisioning decisions to meet the I/O demands of a time-varying workload while avoiding unnecessary costs and Service Level Objective (SLO) violations. We also do a case-study to analyze the practical feasibility of dynamic provisioning and the associated performance effects in a virtualized environment, which forms the basis of our approach. Our approach is able to suggest the optimal provisioning decisions, for a given workload, that minimize cost and meet the SLO. We evaluate the approach using workload data obtained from real systems to demonstrate its cost-effectiveness, sensitivity to various system parameters, and runtime feasibility for use in real systems.

Multi-tenant cloud computing datacenters run diverse workloads, inside virtual machines (VMs), with time varying resource demands. Compute capacity multiplexing systems dynamically manage the placement of VMs on physical machines to ensure that their resource demands are always met while simultaneously optimizing on the total datacenter compute capacity being used. In essence, they give the cloud its fundamental property of being able to dynamically expand and contract resources required on-demand. At large scale datacenters though there are two practical realities that designers of compute capacity multiplexing systems need to deal with: (a) maintaining low operational overhead given variable cost of performing management operations necessary to allocate and multiplex resources, and (b) the prevalence of a large number and wide variety of faults in hardware, software and due to human error, that impair multiplexing efficiency. In this thesis we propound the notion that explicitly designing the methods and abstractions used in capacity multiplexing systems for this reality is critical to better achieve administrator and customer goals at large scales. To this end the thesis makes the following contributions: (i) CCM - a hierarchically organized compute capacity multiplexer that demonstrates that simple designs can be highly effective at multiplexing capacity with low overheads at large scales compared to complex alternatives, (ii) Xerxes - a distributed load generation framework for flexibly and reliably benchmarking compute capacity allocation and multiplexing systems, (iii) A speculative virtualized infrastructure management stack that dynamically replicates management operations on virtualized entities, and a compute capacity multiplexer for this environment, that together provide fault-scalable management performance for a broad class of commonly occurring faults in large scale datacenters. Our systems have been implemented in an industry-strength cloud infrastructure built on top of the VMware vSphere virtualization platform and the popular open source OpenStack cloud computing platform running ESXi and Xen hypervisors, respectively. Our experiments have been conducted in a 700 server datacenter using the Xerxes benchmark replaying trace data from production clusters, simulating parameterized scenarios like flash crowds, and also using a suite of representative cloud applications. Results from these scenarios demonstrate the effectiveness of our design techniques in real-life large scale environments.

Recent years have seen an uptick in master boot record (MBR) based rootkits that load before the Windows operating system and subvert the operating system’s own procedures. As such, MBR rootkits are difficult to counter with operating system-based antivirus software that runs at the same privilege-level as the rookits. Hypervisors operate at a higher privilege level than the guests they manage, creating a high-ground position in the host. This high-ground position can be exploited to perform security checks on the virtual machine guests where the checking software is isolated from guest-based viruses. The efficient introspection system described in this thesis targets existing virtualized systems to improve security with real-time, concurrent memory introspection capabilities. Efficient introspection decouples memory introspection from virtual machine guest execution, establishes coherent and consistent memory views between the host and running guest, while maintaining normal guest operation. Existing introspection systems have provided one or two of these properties but not all three at once. This thesis presents a new concurrent-computing approach – high-performance memory snapshotting – to accelerating hypervisor based introspection of virtual machine guest memory that combines all three elements to improve performance and security. Memory snapshots create a coherent and consistent memory view of the guest that can be shared with the independently running introspection application. Three memory snapshotting mechanisms are presented and evaluated for their impact on normal guest operation. Existing introspection systems and security protection techniques that were previously dismissed as too slow are now be enabled by efficient introspection. This thesis explains why existing introspection systems are inadequate, describes how existing system performance can be improved, evaluates an efficient introspection prototype on both applications and microbenchmarks, and discusses two potential security applications that are enabled by efficient introspection. These applications point to efficient introspection’s utility for supporting useful security applications.

Modern offices are crowded with personal computers. While studies have shown these to be idle most of the time, they remain powered, consuming up to 60% of their peak power. Hardware based solutions engendered by PC vendors (e.g., low power states, Wake-on-LAN) have proven unsuccessful because, in spite of user inactivity, these machines often need to remain network active in support of background applications that maintain network presence. Recent solutions have been proposed that perform consolidation of idle desktop virtual machines. However, desktop VMs are often large requiring gigabytes of memory. Consolidating such VMs, creates large network transfers lasting in the order of minutes, and utilizes server memory inefficiently. When multiple VMs migrate simultaneously, each VM’s experienced migration latency grows, and this limits the use of VM consolidation to environments in which only a few daily migrations are expected per VM. This thesis introduces partial VM migration, an approach that transparently migrates only the working set of an idle VM, by migrating memory pages on-demand. It creates a partial replica of the desktop VM on the consolidation server by copying only VM metadata, and transferring pages to the server, as the VM accesses them. This approach places desktop PCs in low power state when inactive and resumes them to running state when pages are needed by the VM running on the consolidation server. Jettison, our software prototype of partial VM migration for off-the-shelf PCs, can deliver 78% to 91% energy savings during idle periods lasting more than an hour, while providing low migration latencies of about 4 seconds, and migrating minimal state that is under an order of magnitude of the VM’s memory footprint. In shorter idle periods of up to thirty minutes, Jettison delivers savings of 7% to 31%. We present two approaches that increase energy savings attained with partial VM migration, especially in short idle periods. The first, Context-Aware Selective Resume, expedites PC resume and suspend cycle times by supplying a context identifier at desktop resume, and initializing only devices and code that are relevant to the context. CAESAR, the Context-Aware Selective Resume framework, enables applications to register context vectors that are invoked when the desktop is resumed with matching context. CAESAR increases energy savings in short periods of five minutes to an hour by up to 66%. The second approach, the low power page cache, embeds network accessible low power hardware in the PC, to enable serving of pages to the consolidation server, while the PC is in low power state. We show that Oasis, our prototype page cache, addresses the shortcomings of energy-oriented on-demand page migration by increasing energy savings, especially during short idle periods. In periods of up to an hour, Oasis increases savings by up to twenty times.

Modern offices are crowded with personal computers. While studies have shown these to be idle most of the time, they remain powered, consuming up to 60% of their peak power. Hardware based solutions engendered by PC vendors (e.g., low power states, Wake-on-LAN) have proven unsuccessful because, in spite of user inactivity, these machines often need to remain network active in support of background applications that maintain network presence. Recent solutions have been proposed that perform consolidation of idle desktop virtual machines. However, desktop VMs are often large requiring gigabytes of memory. Consolidating such VMs, creates large network transfers lasting in the order of minutes, and utilizes server memory inefficiently. When multiple VMs migrate simultaneously, each VM’s experienced migration latency grows, and this limits the use of VM consolidation to environments in which only a few daily migrations are expected per VM. This thesis introduces partial VM migration, an approach that transparently migrates only the working set of an idle VM, by migrating memory pages on-demand. It creates a partial replica of the desktop VM on the consolidation server by copying only VM metadata, and transferring pages to the server, as the VM accesses them. This approach places desktop PCs in low power state when inactive and resumes them to running state when pages are needed by the VM running on the consolidation server. Jettison, our software prototype of partial VM migration for off-the-shelf PCs, can deliver 78% to 91% energy savings during idle periods lasting more than an hour, while providing low migration latencies of about 4 seconds, and migrating minimal state that is under an order of magnitude of the VM’s memory footprint. In shorter idle periods of up to thirty minutes, Jettison delivers savings of 7% to 31%. We present two approaches that increase energy savings attained with partial VM migration, especially in short idle periods. The first, Context-Aware Selective Resume, expedites PC resume and suspend cycle times by supplying a context identifier at desktop resume, and initializing only devices and code that are relevant to the context. CAESAR, the Context-Aware Selective Resume framework, enables applications to register context vectors that are invoked when the desktop is resumed with matching context. CAESAR increases energy savings in short periods of five minutes to an hour by up to 66%. The second approach, the low power page cache, embeds network accessible low power hardware in the PC, to enable serving of pages to the consolidation server, while the PC is in low power state. We show that Oasis, our prototype page cache, addresses the shortcomings of energy-oriented on-demand page migration by increasing energy savings, especially during short idle periods. In periods of up to an hour, Oasis increases savings by up to twenty times.

Η παρούσα διπλωματική έχει σαν σκοπό τη διερεύνηση της τεχνολογίας του Cloud Computing και της τεχνολογίας της Virtualization που την στηρίζει. Παρουσίαση της ιστορίας και μια τεχνική παρουσίαση των δυνατοτήτων και των καταβολών των τεχνολογιών. Αναφέρονται πρακτικές εφαρμογές που μπορούν οι συγκεκριμένες τεχνολογίες να εφαρμοστούν και τους σκοπούς που θα εξυπηρετήσουν. Επίσης γίνεται μια πιο αναλυτική παρουσίαση δυο προγραμμάτων (Xen Hypervisor – για το επίπεδο της Virtualization, Eucalyptus – σαν πλατφόρμα για τη δημιουργία IaaS Clouds). Παρουσιάζονται επίσης σύντομοι οδηγοί για την εγκατάσταση ενός Cloud, καθώς και το configuration μαζί με τους λόγους που χρησιμοποιήθηκε. / The scope of this thesis is to study the technology of Cloud Computing and the Virtualization technology that is supporting it. A presentation of the history, a technical overview and the origins of these technologies. There are mentioned some fields that the specified technologies could apply and the purposes that they would serve. On the third chapter, a more detailed presentation of two pieces of software is given (Xen Hypervisor – for the Virtualization Layer, Eucalyptus – as the platform to create IaaS Clouds). In the end quick how-to guides are described on the procedure to install a Cloud, the configuration and the reasons of the specific set up as well.

New and innovative technologies to improve the techniques that are already being used are constantly developing. This project was about evaluating if containers could be something for the IT company Tieto to use on their products in telecommunications. Container are portable, standalone, executable lightweight packages of software that also contains all it needs to run the software. Containers are a very hot topic right now and are a fast-growing technology. Tieto wanted an investigation of the technology and it would be carried out with certain requirements where the main requirement was to have a working and executable protocol stack in a container environment. In the investigation, a proof of concept was developed, proof of concept is a realization of a certain method or idea in order to demonstrate its feasibility. The proof of concept led to Tieto wanting additional experiments carried out on containers. The experiments investigated if equal performance could be achieved with containers compared to the method with virtual machine used by Tieto today. The experiment observed a small performance reduction of efficiency, but it also showed benefits such as higher flexibility. Further development of the container method could provide a just as good and equitable solution. The project can therefore be seen as successful whereas the proof of concept developed, and experiments carried out both points to that this new technology will be part of Tieto's product development in the future.

AN ABSTRACT OF THE THESIS OF Abdulaziz M. AlNowiser, for the Master of Science degree in Computer Science, presented on November 1, 2013, at Southern Illinois University Carbondale. TITLE: TOWARD ENERGY-EFFICIENT SCHEDULING USING WEIGHTED ROUND- ROBIN AND VM REUSE MAJOR PROFESSOR: Dr. Michelle M. Zhu In recent years, the rapid evolving Cloud Computing technologies multiply challenges such as minimizing power consumption and meeting Quality-of-Services (QoS) requirements in the presence of heavy workloads from a large number of users using shared computing resources. Powering a middle-sized data center normally consumes 80,000kW power every year and computer servers consume around .5% of the global power [1]. Statistics for 5000 production servers over a six-month period show that only 10-50% of the total capacity has been effectively used, and a large portion of the resources is actually wasted. In order to address the skyrocket energy cost from the high level resource management aspect, we propose an energy efficient job scheduling approach based on a modified version of Weighted Round Robin scheduler that incorporates VMs reuse and live VM migration without compromising the Service Level Agreement (SLA). The Weighted Round Robin scheduler can monitor the running VMs status for possible VM sharing for job consolidation or migration. In addition, the VMs utilization rate is observed to start live migration from the over-utilizing Processing Element (PE) to under-utilized PEs or to the hibernated PEs by sending WOL (Wake-On-LAN) signal to activate them. The simulation experiments are conducted under the CloudReports environment based on open source CloudSim simulator. The comparisons with other similar scheduling algorithms demonstrate that our enhanced Weighted Round Robin algorithm (EWRR) can achieve considerable better performance in terms of energy consumption and resource utilization rate.

Confrontées à l'augmentation des coûts de mise en place et de maintenance des systèmes informatiques, les entreprises se tournent vers des solutions d'externalisation telles que le Cloud Computing. Le Cloud se basent sur la virtualisation comme principale technologie permettant la mutualisation. L'utilisation de la virtualisation apporte de nombreux défis donc les principaux portent sur les performances des applications dans les machines virtuelles (VM) et la prévisibilité de ces performances. Dans un système virtualisé, les ressources matérielles sont partagées entre toutes les VMs du système. Dans le cas du CPU, c'est l'ordonnanceur de l'hyperviseur qui se charge de le partager entre tous les processeurs virtuels (vCPU) des VMs. L'hyperviseur réalise une allocation à temps partagé du CPU entre tous les vCPUs des VMs. Chaque vCPU a accès au CPU périodiquement. Ainsi, les vCPUs des VMs n'ont pas accès de façon continue au CPU, mais plutôt discontinue. Cette discontinuité est à l'origine de nombreux problèmes sur des mécanismes tels que la gestion d'interruption et les mécanismes de synchronisation de bas niveau dans les OS invités. Dans cette thèse, nous proposons deux contributions pour répondre à ces problèmes dans la virtualisation. La première est un nouvel ordonnanceur de l'hyperviseur qui adapte dynamiquement la valeur du quantum dans l'hyperviseur en fonction du type des applications dans les VMs sur une plate-forme multi-coeurs. La seconde contribution est une nouvelle primitive de synchronisation (nommée I-Spinlock) dans l'OS invité. Dans un Cloud fournissant un service du type IaaS, la VM est l'unité d'allocation. Le fournisseur établit un catalogue des types de VMs présentant les différentes quantités de ressources qui sont allouées à la VM vis-à-vis des différents périphériques. Ces ressources allouées à la VM correspondent à un contrat sur une qualité de service négocié par le client auprès du fournisseur. L'imprévisibilité des performances est la conséquence de l'incapacité du fournisseur à garantir cette qualité de service. Deux principales causes sont à l'origine de ce problème dans le Cloud: (i) un mauvais partage des ressources entre les différentes VMs et (ii) l'hétérogénéité des infrastructures dans les centres d'hébergement. Dans cette thèse, nous proposons deux contributions pour répondre au problème d'imprévisibilité des performances. La première contribution s'intéresse au partage de la ressource logicielle responsable de la gestion des pilotes, et propose une approche de facturation du temps CPU utilisé par cette couche logiciel aux VMs. La deuxième contribution s'intéresse à l'allocation du CPU dans les Clouds hétérogènes. Dans cette contribution, nous proposons une approche d'allocation permettant de garantir la capacité de calcul allouée à une VM quelle que soit l'hétérogénéité des CPUs dans l'infrastructure. / As a reaction to the increasing costs of setting up and maintaining IT systems, companies are turning to solutions such as Cloud Computing. Cloud computing is based on virtualization as the main technology for mutualisation. The use of virtualization brings many challenges. The main ones concern the performance of the applications in the virtual machines (VM) and the predictability of these performances. In a virtualized system, hardware resources are shared among all VMs in the system. In the case of the CPU, it is the scheduler of the hypervisor that is in charge of sharing the CPU among all the virtual processors (vCPU) of the VMs. The hypervisor uses a time-sharing approach to allocate the CPU. Each vCPU has access to the CPU periodically. Thus, the vCPU of the VMs do not have continuous access to the CPU, but rather discontinuous. This discontinuity is causing many problems on mechanisms such as interuption handling and low-level synchronization mechanisms in guest OSs. In this thesis, we propose two contributions to address these problems in virtualization. The first is a new hypervisor scheduler that dynamically adapts the quantum value in the hypervisor according to the type of applications in the VMs on a multi-core platform. The second contribution is a new synchronization primitive (named I-Spinlock) in the guest OS. In a cloud providing a service of the IaaS type, the VM is the allocation unit. The provider establishes a catalogue presenting the different quantities of resources that are allocated to the VM regarding various devices. These resources allocated to the VM correspond to a contract on a quality of service negotiated by the customer with the provider. The unpredictability of performance is the consequence of the incapability of the provider to guarantee this quality of service. There are two main causes of this problem in the Cloud: (i) poor resource sharing between different VMs and (ii) heterogeneity of infrastructure in hosting centers. In this thesis, we propose two contributions to answer the problem of performance unpredictability. The first contribution focuses on the sharing of the software resource responsible for managing the drivers, and proposes to bill the CPU time used by this software layer to VMs. The second contribution focuses on the allocation of the CPU in heterogeneous clouds. In this contribution, we propose an allocation approach to guarantee the computing capacity allocated to a VM regardless of the heterogeneity of the CPUs in the infrastructure.

abstract: With the software-defined networking trend growing, several network virtualization controllers have been developed in recent years. These controllers, also called network hypervisors, attempt to manage physical SDN based networks so that multiple tenants can safely share the same forwarding plane hardware without risk of being affected by or affecting other tenants. However, many areas remain unexplored by current network hypervisor implementations. This thesis presents and evaluates some of the features offered by network hypervisors, such as full header space availability, isolation, and transparent traffic forwarding capabilities for tenants. Flow setup time and throughput are also measured and compared among different network hypervisors. Three different network hypervisors are evaluated: FlowVisor, VeRTIGO and OpenVirteX. These virtualization tools are assessed with experiments conducted on three different testbeds: an emulated Mininet scenario, a physical single-switch testbed, and also a remote GENI testbed. The results indicate that network hypervisors bring SDN flexibility to network virtualization, making it easier for network administrators to define with precision how the network is sliced and divided among tenants. This increased flexibility, however, may come with the cost of decreased performance, and also brings additional risks of interoperability due to a lack of standardization of virtualization methods. / Dissertation/Thesis / Masters Thesis Engineering 2016