Selecting the correct inter-process communication method isan important aspect of ensuring effective inter-vm and inter-container process communication. We will conduct a study ofIPC methods which might be useful and fits the Qemu/KVMvirtual machine and Docker container environments, and se-lect those that fit our criteria. After implementing our chosenmethods we will benchmark them in a test suite to find theones with highest performance in terms of speed. Our resultsshow that, at the most common message sizes, Unix DomainSockets work best for containers and Transparent Inter Pro-cess Communication has the best performance between vir-tual machines out of the chosen methods.

Virtualization is the foundation on which cloud services build their business. It supports the infrastructure for the largest companies around the globe and is a key component for scaling software for the ever-growing technology industry. If companies decide to use virtualization as part of their infrastructure it is important for them to quickly and reliably have a way to choose a virtualization technology and tweak the performance of that technology to fit their intended usage. Unfortunately, while many papers exist discussing and testing the performance of various virtualization systems, most of these performance tests do not take into account components that can be configured to improve performance for certain scenarios. This study provides a comparison of how three hypervisors (VMWare vSphere, Citrix XenServer, and KVM) perform under different sets of configurations at this point and which system workloads would be ideal for these configurations. This study also provides a means in which to compare different configurations with each other so that implementers of these technologies have a way in which to make informed decisions on which components should be enabled for their current or future systems.

FundaÃÃo Cearense de Apoio ao Desenvolvimento Cientifico e TecnolÃgico / O escalonamento de recursos à um processo chave para a plataforma de ComputaÃÃo em Nuvem, que geralmente utiliza mÃquinas virtuais (MVs) como unidades de escalonamento. O uso de tÃcnicas de virtualizaÃÃo fornece grande flexibilidade com a habilidade de instanciar vÃrias MVs em uma mesma mÃquina fÃsica (MF), modificar a capacidade das MVs e migrÃ-las entre as MFs. As tÃcnicas de consolidaÃÃo e alocaÃÃo dinÃmica de MVs tÃm tratado o impacto da sua utilizaÃÃo como uma medida independente de localizaÃÃo. à geralmente aceito que o desempenho de uma MV serà o mesmo, independentemente da MF em que ela à alocada. Esta à uma suposiÃÃo razoÃvel para um ambiente homogÃneo, onde as MFs sÃo idÃnticas e as MVs estÃo executando o mesmo sistema operacional e aplicativos. No entanto, em um ambiente de ComputaÃÃo em Nuvem, espera-se compartilhar um conjunto composto por recursos heterogÃneos, onde as MFs podem variar em termos de capacidades de seus recursos e afinidades de dados. O objetivo principal deste trabalho à apresentar uma arquitetura que possibilite a padronizaÃÃo da representaÃÃo do poder de processamento das MFs e MVs, em funÃÃo de Unidades de Processamento (UPs), apoiando-se na limitaÃÃo do uso da CPU para prover isolamento de desempenho e manter a capacidade de processamento das MVs independente da MF subjacente. Este trabalho busca suprir a necessidade de uma soluÃÃo que considere a heterogeneidade das MFs presentes na infraestrutura da Nuvem e apresenta polÃticas de escalonamento baseadas na utilizaÃÃo das UPs. A arquitetura proposta, chamada FairCPU, foi implementada para trabalhar com os hipervisores KVM e Xen, e foi incorporada a uma nuvem privada, construÃda com o middleware OpenNebula, onde diversos experimentos foram realizados para avaliar a soluÃÃo proposta. Os resultados comprovam a eficiÃncia da arquitetura FairCPU em utilizar as UPs para reduzir a variabilidade no desempenho das MVs, bem como para prover uma nova maneira de representar e gerenciar o poder de processamento das MVs e MFs da infraestrutura. / Resource scheduling is a key process for cloud computing platform, which generally uses virtual machines (VMs) as scheduling units. The use of virtualization techniques provides great flexibility with the ability to instantiate multiple VMs on one physical machine (PM), migrate them between the PMs and dynamically scale VMâs resources. The techniques of consolidation and dynamic allocation of VMs have addressed the impact of its use as an independent measure of location. It is generally accepted that the performance of a VM will be the same regardless of which PM it is allocated. This assumption is reasonable for a homogeneous environment where the PMs are identical and the VMs are running the same operating system and applications. Nevertheless, in a cloud computing environment, we expect that a set of heterogeneous resources will be shared, where PMs will face changes both in terms of their resource capacities and as also in data affinities. The main objective of this work is to propose an architecture to standardize the representation of the processing power by using processing units (PUs). Adding to that, the limitation of CPU usage is used to provide performance isolation and maintain the VMâs processing power at the same level regardless the underlying PM. The proposed solution considers the PMs heterogeneity present in the cloud infrastructure and provides scheduling policies based on PUs. The proposed architecture is called FairCPU and was implemented to work with KVM and Xen hypervisors. As study case, it was incorporated into a private cloud, built with the middleware OpenNebula, where several experiments were conducted. The results prove the efficiency of FairCPU architecture to use PUs to reduce VMsâ performance variability, as well as to provide a new way to represent and manage the processing power of the infrastructureâs physical and virtual machines.

Existuje veľa rôznych dôvodov pre spoločnosti a organizácie, prečo by mali investovať do virtualizácie. Asi najväčší dôvod je finančná motivácia, pretože nasadenie virtualizácie môže ušetriť nemálo peňazí. Táto práca sa zaoberá práve problémom virtualizácie I/O operácií v sieťovom prostredí. Cieľom práce je tvorba softvérových ovládačov pre I/O virtualizáciu, ktoré by mohli pracovať s hardvérovo akcelerovanými sieťovými kartami. Hlavným prínosom ovládačov by mala byť použiteľnosť a čo najmenšia strata prenosového výkonu vo virtualizovanom prostredí. Pred popisom finálnych detailov ovládačov je však potrebné uviesť potrebné teoretické základy. Teoretická časť sa zaoberá súčasnými trendami vo virtualizácii I/O, technológiami ako sú virtio, vhost, SR-IOV, VFIO a mdev. V praktickej časti sú navrhuté dva spôsoby riešenia problému. Prvým je použitie technológie virtio (emulácia softvéru). Druhé je založené na technológii VFIO-mdev (hybridná paravirtualizácia). Pokiaľ sa jedná o výkon a konfigurovateľnosť zariadení, oba prístupy majú rôzne benefity. Tieto riešenia majú aj svoje nevýhody, ako je zložitosť riešenia a náročnosť integrácie do systému. Požadované ciele boli úspešne dosiahnuté vo forme prototypu ovládača nfb_mdev.

The goal of this diploma thesis is to design the user interface and implement the information system as a web application. Using the custom implemented library the system communicates with GraphQL server which manages the client data. The thesis describes possible solutions for physical servers automatization. The application provides the application interface to manage virtual servers. Automatization is possible without human interaction. Connection to the virtualization technologies is handled by web interface APIs or custom scripts running in the virtual system terminal. There is a monitoring system built over project components. The thesis also describes the continuous integration using Gitlab tools. Running the configuration task is solved using the Unix CRON system.