Virtualization technology has attracted considerable interest. It allows several virtual machines to run concurrently inside a physical host, which brings multiple advantages. One of the most useful features is called live migration, during which a virtual machine can be migrated over network with minimal disruption. So far, most existing migration algorithms are focused on transferring the runtime state over high-speed, low-delay network. They all require shared storage for file systems. However, this sharing sometimes becomes impossible because of performance loss. Thus, the whole system needs to be transferred during migration. In this thesis, we introduce a Virtual Machine Management System which contains a block-level solution. Combined with pre-copying the runtime state, we can migrate an entire virtual machine over low-bandwidth, high-delay network with minimum service downtime. We show that this is sufficient even for interactive workloads. We also provide snapshots and full backup for the virtual machine.

No description available.

Cloud computing has grown in prevalence from recent years due to its concept of computing as a service, thereby, allowing users to offload the infrastructure management costs and tasks to a cloud provider. Cloud providers leverage server virtualization technology for efficient resource utilization, faster provisioning times, reduced energy consumption, etc. Cloud computing inherits a key feature of server virtualization which is the live migration of virtual machines (VMs). This technique allows transferring of a VM from one host to another with minimal service interruption. However, live migration is a complex process and with a cloud management software used by cloud providers for management, there could be a significant influence on the migration process. This thesis work aims to investigate the complex process of live migration performed by the hypervisor as well as the additional steps involved when a cloud management software or platform is present and form a timeline of these collection of steps or phases. The work also aims to investigate the performance of these phases, in terms of time, when migrating VMs with different sizes and workloads. For this thesis, the Kernel-based Virtual Machine (KVM) hypervisor and the OpenStack cloud software have been considered. The methodology employed is experimental and quantitative. The essence of this work is investigation by network passive measurements. To elaborate, this thesis work performs migrations on physical test-beds and uses measurements to investigate and evaluate the migration process performed by the KVM hypervisor as well as the OpenStack platform deployed on KVM hypervisors. Experiments are designed and conducted based on the objectives to be met. The results of the work primarily include the timeline of the migration phases of both the KVM hypervisor and the OpenStack platform. Results also include the time taken by each migration phase as well as the total migration time and the VM downtime. The results indicate that the total migration time, downtime and few of the phases increase with increase in CPU load and VM size. However, some of the phases do not portray any such trend. It has also been observed that the transfer stage alone does not contribute and influence the total time but every phase of the process has significant influence on the migration process. The conclusions from this work is that although a cloud management software aids in managing the infrastructure, it has notable impact on the migration process carried out by the hypervisor. Moreover, the migration phases and their proportions not only depend on the VM but on the physical environment as well. This thesis work focuses solely on the time factor of each phase. Further evaluation of each phase with respect to its resource utilization can provide better insight into probable optimization opportunities.

Virtualization is a key technology for cloud computing as it allows several operating system instances to run on the same machine, enhances resource manageability and enables flexible definition of billing units. Virtualization works by adding a software layer, a hypervisor, on top of the hardware platform. Virtual Machines, \emph{VMs}, are run on top of the hypervisor, which provisions hardwares resources to the VM guests. In addition to enabling higher utilization of hardware resources, the ability to move VMs from one host to another is an important feature. Live migration is the concept of migrating a VM while it is running and responding to requests. Since VMs can be re-located while running, live migration allows for better hardware utilization. This is because placement of services can be performed dynamically and not only when the are started. Live migration is also a useful tool for administrative purposes. If a server needs to be taken off-line for maintenance reasons, it can be cleared of services by live migrating these to other hosts. This thesis investigates the principles behind live migration. The common live migration approaches in use today are evaluated and common objectives are presented as well as challenges that have to be overcome in order to implement an ideal live migration algorithm. The performance of common live migration approaches is also evaluated and it is found that even though live migration is supported by most hypervisors, it has drawbacks which makes the technique hard to use in certain situations. Migrating CPU and/or memory intensive VMs or migrating VMs over low-bandwidth links is a problem regardless of which approach that is used. To tackle this problem, two improvements to live migration are proposed and evaluated, delta compression and dynamic page transfer reordering. Both improvements demonstrate better performance than the standard algorithm when migrating CPU and/or memory intensive VMs and migrating over low bandwidth links. Finally, recommendations are made on which live migration approach to use depending on the scenario and also what improvements to the standard live migration algorithms should be used and when.

A key aspect of cloud computing is the promise of infinite, scalable resources, and that cloud services should scale up and down on demand. This thesis investigates methods for dynamic resource allocation and management of services in cloud datacenters, introducing new approaches as well as improvements to established technologies.Virtualization is a key technology for cloud computing as it allows several operating system instances to run on the same Physical Machine, PM, and cloud services normally consists of a number of Virtual Machines, VMs, that are hosted on PMs. In this thesis, a novel virtualization approach is presented. Instead of running each PM isolated, resources from multiple PMs in the datacenter are disaggregated and exposed to the VMs as pools of CPU, I/O and memory resources. VMs are provisioned by using the right amount of resources from each pool, thereby enabling both larger VMs than any single PM can host as well as VMs with tailor-made specifications for their application. Another important aspect of virtualization is live migration of VMs, which is the concept moving VMs between PMs without interruption in service. Live migration allows for better PM utilization and is also useful for administrative purposes. In the thesis, two improvements to the standard live migration algorithm are presented, delta compression and page transfer reordering. The improvements can reduce migration downtime, i.e., the time that the VM is unavailable, as well as the total migration time. Postcopy migration, where the VM is resumed on the destination before the memory content is transferred is also studied. Both userspace and in-kernel postcopy algorithms are evaluated in an in-depth study of live migration principles and performance.Efficient mapping of VMs onto PMs is a key problem for cloud providers as PM utilization directly impacts revenue. When services are accepted into a datacenter, a decision is made on which PM should host the service VMs. This thesis presents a general approach for service scheduling that allows for the same scheduling software to be used across multiple cloud architectures. A number of scheduling algorithms to optimize objectives like revenue or utilization are also studied. Finally, an approach for continuous datacenter consolidation is presented. As VM workloads fluctuate and server availability varies any initial mapping is bound to become suboptimal over time. The continuous datacenter consolidation approach adjusts this VM-to-PM mapping during operation based on combinations of management actions, like suspending/resuming PMs, live migrating VMs, and suspending/resuming VMs. Proof-of-concept software and a set of algorithms that allows cloud providers to continuously optimize their server resources are presented in the thesis.

Context: Cloud computing is one of the most widely used technologies all over the world that provides numerous products and IT services. Virtualization is one of the innovative technologies in cloud computing which has advantages of improved resource utilisation and management. Live migration is an innovative feature of virtualization that allows a virtual machine or container to be transferred from one physical server to another.  Live migration is a complex process which can have a significant impact on cloud computing when used by the cloud-based software.  Objectives: In this study, live migration of LXC and OpenVZ containers has been performed.  Later the performance of LXC and OpenVZ has been conducted in terms of total migration time and downtime. Further CPU utilisation, disk utilisation and an average load of the servers is also evaluated during the process of live migration. The main aim of this research is to compare the performance of LXC and OpenVZ during live migration of containers.  Methods: A literature study has been done to gain knowledge about the process of live migration and the metrics that are required to compare the performance of LXC and OpenVZ during live migration of containers. Further an experiment has been conducted to compute and evaluate the performance metrics that have been identified in the literature study. The experiment was done to investigate and evaluate migration process for both LXC and OpenVZ. Experiments were designed and conducted based on the objectives which were to be met. Results:  The results of the experiments include the migration performance of both LXC and OpenVZ. The performance metrics identified in the literature review, total migration time and downtime, were evaluated for LXC and OpenVZ. Further graphs were plotted for the CPU utilisation, disk utilisation, and average load during the live migration of containers. The results were analysed to compare the performance differences between OpenVZ and LXC during live migration of containers. Conclusions.  The conclusions that can be drawn from the experiment. LXC has shown higher utilisation, thus lower performance when compared with OpenVZ. However, LXC has less migration time and downtime when compared to OpenVZ.

Migrer à chaud une machine virtuelle (VM) est une opération basique dans un centre de données. Tous les jours, des VM sont migrées pour répartir la charge, économiser de l'énergie ou préparer la maintenance de serveurs. Bien que les problèmes de placement des VM soient beaucoup étudiés, on observe que la gestion des migrations permettant de transiter vers ces nouveaux placements reste un domaine de second plan. Cette phase est cependant critique car chaque migration à un coût en terme de CPU, de bande passante et d'énergie. Des algorithmes de décision reposent alors sur des hypothèses irréalistes et calculent des ordonnancements conduisant à des migrations longues et incontrôlables qui réduisent les bénéfices attendus de la ré-organisation des VM.Dans cette thèse nous nous sommes fixé comme objectif d'améliorer la qualité des ordonnancements de migrations dans les centres de données. Pour cela, nous avons d'abord modélisé l'ordonnancement de migrations en considérant l'architecture réseau et l'activité mémoire des VM. Pour évaluer l'efficacité de notre modèle, nous avons ensuite implémenté un ordonnanceur de migrations au sein du gestionnaire de VM BtrPlace. Nous avons ensuite étendu notre ordonnanceur en développant des contraintes d'ordonnancement, des objectifs personnalisés, une heuristique de recherche ainsi qu'un modèle énergétique.Nous avons validé notre approche par l'étude pratique de scénarios d'ordonnancement réalisés en environnement réel. Nous avons ainsi pu analyser la précision de notre modèle de migration, valider la qualité des décisions prises par notre modèle d'ordonnancement et évaluer l'extensibilité ainsi que le passage à l'échelle de notre solution / A live-migration of a virtual machine (VM) is a basic operation in a data center. Every day, VMs are migrated to distribute the load, save energy or prepare maintenance operations on production servers. Although VM placement problems have been extensively studied, we observe that the migrations management needed to apply these new placements did not get much attention. This phase is however critical as each migration has a cost in terms of CPU, bandwidth and energy. Decision algorithms are thus based on unrealistic assumptions and compute schedules which can lead to unnecessarily long and uncontrollable migrations. This reduces the ultimate benefits expected from the VMs re-organization.In this thesis, our main ojective is to improve the efficiency of live-migrations scheduling within data centers. To achieve our goal, we have first modeled the scheduling of live migrations based on the network architecture and the VMs memory activity. To evaluate the efficiency of our model, we have then implemented and optimized a migrations scheduler within the VMs manager BtrPlace. We have then extended our scheduler by developing scheduling constraints, custom objectives, a search heuristic and an energy model.We have validated our approach by the practical study of many scheduling scenarios executed in a real environment. We have then analyzed the accuracy of our migration model, assessed the quality of the decisions taken by our scheduling model, and evaluated the extensibility and the scalability of our solution

Historically virtual machines have been the backbone of the cloud-industry, allowing cloud-providers to offer virtualized multi-tenant solutions. A key aspect of the cloud is its flexibility and abstraction of the underlying hardware. Virtual machines can enhance this aspect by enabling support for live migration and failover. Live migration is the process of moving a running virtual machine from one host to another and failover ensures that a failed virtual machine will automatically be restarted (possibly on another host). Today, as containers continue to increase in popularity and make up a larger portion of the cloud, often replacing virtual machines, it becomes increasingly important for these processes to be available to containers as well. However, little support for container live migration and failover exists and remains largely experimental. Furthermore, no solution seems to exists that offers both live migration and failover for containers in a unified solution. The thesis presents a proof-of-concept implementation and description of a system that enables support for both live migration and failover for containers by extending current container migration techniques. It is able to offer this to any OCI-compliant container, and could therefore potentially be integrated into current container and container orchestration frameworks. In addition, measurements for the proof-of-concept implementation are provided and used to compare the proof-of-concept implementation to a current container migration technique. Furthermore, the thesis presents an overview of the history and implementation of containers, current migration techniques, and metrics that can be used for measuring different migration techniques are introduced. The paper concludes that current container migration techniques can be extended in order to support both live migration and failover, and that in doing so one might expect to achieve a downtime equal to, and total migration time lower than that of pre-copy migration. Supporting both live migration and failover, however, comes at a cost of an increased amount of data needed to be transferred between the hosts.

The slowdown of Moore's Law opened a new era of computer research and development. Researchers started exploring alternatives to the traditional CPU design. A constant increase in consumer demands led to the development of CMPs, GPUs, and FPGAs. Recent research proposed the development of heterogeneous-ISA systems and implemented the necessary systems software to make such systems functional. Evaluations have shown that heterogeneous-ISA systems can offer better throughput and energy efficiency than homogeneous-ISA systems. Due to their low cost, ARM servers are now being adopted in data centers (e.g., AWS Graviton). While prior work provided the infrastructure necessary to run applications on heterogeneous-ISA systems, their dependency on a specialized kernel and a custom compiler increases deployment and maintenance costs. This thesis presents HetMigrate, a framework to live-migrate Linux processes over heterogeneous-ISA systems. HetMigrate integrates with CRIU, a Linux mechanism for process migration, and runs on stock Linux operating systems which improves its deployability. Furthermore, HetMigrate transforms the process's state externally without instrumenting state transformation code into the process binaries which has security benefits and also improves deployability. Our evaluations on Redis server and NAS Parallel Benchmarks show that HetMigrate takes an average of 720ms to fully migrate a process across ISAs while maintaining its state. Moreover, live-migrating with HetMigrate reduces the attack surface of a process by up to 72.8% compared to prior work. Additionally, HetMigrate is easier to deploy in real-world systems compared to prior work. To prove the deployability we ran HetMigrate on a variety of environments like cloud instances (e.g. Cloud Lab), local setups virtualized with QEMU/KVM, and a server-embedded board pair. Similar to works in the past, we also evaluated the energy and throughput benefits that heterogeneous-ISA systems can offer by connecting a Xeon server to three embedded boards over the network. We observed that selectively offloading compute-intensive workloads to embedded boards can increase energy efficiency by up to 39% and throughput by up to 52% while increasing the cost by just 10%. / Master of Science / In 1965 Gordon Moore predicted that the number of transistors in a chip will double every two years. Commonly referred to as "Moore's Law" it no longer holds true and its slowdown opened a new era of computer research and development. Researchers started exploring alternatives to traditional computer designs. A constant increase in consumer demands led to the development of portable, faster, and cheaper computers. Some researchers also started exploring the performance and energy benefits of computing systems that had heterogeneous architecture. Instruction Set Architecture (ISA) is the interface between software and hardware. Recent research proposed the development of systems that had cores of different ISA and implemented the necessary software to make such systems functional. Evaluations have shown that heterogeneous-ISA systems can offer better throughput and energy efficiency than traditional systems. To decrease their cost-to-performance ratio data centers have started adopting servers belonging to diverse architectures making them heterogeneous in nature. While prior work provided the infrastructure necessary to run applications on heterogeneous systems, it suffered from deployability limitations. This thesis presents HetMigrate, a framework that enables stateful program migration in heterogeneous systems. HetMigrate runs on stock open-source operating systems which makes it easy to deploy. Our evaluations show that while HetMigrate performs two orders of magnitude slower than prior work, it can be deployed with ease.

With advancements in the mobile network architecture, from the Fourth Generation to the Fifth Generation, a vast number of new use cases becomes available. Many use cases require cloud-based services, where a service is deployed close to the user. For a user to communicate with a service, it connects to the mobile network base station, Fifth Generation Core network and then to the service. When the user changes physical location, the mobile network and the service must apply mobility techniques. This is to prevent tromboned traffic and provide low latency between user and service. When a handover occurs, so that a user’s attachment point to the mobile network is changed from the one base station to another and the User Plane Function changes, the cloud-based service may have to seamlessly move from one cloud to another as well. In this thesis, a Service mobility framework is proposed and implemented, which enables service live migration between edge clouds and it provides simple RESTful APIs. The evaluation of the framework shows that the proposed implementation adds low delays to the total migration time and the service downtime is also shown to be low in the case of video streaming with no service interruption. / Med framsteg i det mobila nätverkets arkitektur, sett från den Fjärde Generationen till den Femte Generationen, så blir nya användningsområden tillgängliga. Bland de nya användningsområdena inkluderas molnbaserade tjänster, där tjänster är placerade nära användare, dessutom har vissa områden behov av dessa molnbaserade tjänster. För att en användare ska kunna kommunicera med en tjänst så måste den först ansluta till det mobila nätverkets basstationer och sedan till Femte Generationens kärnnätverk, för att sedan kunna kommunicera med tjänsten. När användaren förflyttar sig från en plats till en annan, så måste det mobila nätverket och tjänsten tillämpa rörlighetstekniker, som förflyttning av tjänsten. Förflyttningen är för att förhindra trombonerad trafik och att förse låg latens mellan användare och tjänst. När en överlämning sker, d.v.s att en användares kopplingspunkt till det mobila nätverket ändras, från en basstation till en annan, och att User Plane Function ändras, så kan även den molnbaserade tjänsten förflytta sig sömlöst från ett moln till ett annat. I denna avhandling presenteras ett tjänströrlighetsramverk som möjliggör tjänströrlighet mellan moln och erbjuder enkla RESTfulla API:er. Evaluering av ramverket visar att implementationen bidrar med låga fördröjningar till den totala migrations tiden samt att tjänster med videoströmming har lågt driftstopp utan tjänstavbrott.