Garantia de qualidade de serviço (QOS) em arquitecturas MPLS

Os prestadores de serviços de telecomunicações e operadores de telecomunicações deparam-se com um aumento exponencial das necessidades de largura de banda. A evolução e massificação dos serviços Internet e Intranet pelos serviços públicos e privados deixaram de ser uma mera adaptação do protocolo TCP, à qualidade da ligação sendo uma necessidade a diferenciação do tráfego. As metodologias que asseguram uma qualidade de serviço no âmbito dos fornecedores de serviços internet são a forma de garantir uma qualidade de serviço adequada a cada tipo de tráfego. Estas metodologias são suportadas pela rede IP MPLS dos diversos operadores de telecomunicações no transporte dos diversos serviços dos seus clientes empresarias e domésticos no acesso à internet dos diversos serviços públicos de dados e voz e nas redes virtuais privadas. Os portais aplicacionais são a interface directa com o cliente para definição dos acordos de nível de serviço “Service Level Agreements” e a sua associação à especificação dos níveis de serviço “Service Level Specification”, para posterior relação com a definição de métricas adequadas à qualidade de serviço acordada com o cliente no desenho dos serviços de uma rede IP “MultiProtocol Label Switch”. A proposta consiste em criar uma metodologia para mapear as necessidades de serviços dos clientes em SLAs e registá-los numa base de dados, separando claramente a qualidade do serviço vista na óptica do operador em: arquitectura de rede de transporte, arquitectura do serviço e arquitectura de monitoria. Estes dados são mapeados em parâmetros e especificações de implementação dos serviços de suporte ao negócio do operador tendo em vista a criação de um “Work Flow” fim a fim. Paralelamente define-se os serviços a disponibilizar comercialmente, o conjunto de serviços suportados pela rede e tecnologia IP MPLS com a parametrização de ”Quality of Service Assurance” adequada a cada um, cria-se uma arquitectura de rede de suporte ao transporte base entre os diversos equipamentos agregadores de acessos através do “Backbone”, define-se uma arquitectura de suporte para cada tipo de serviço independente da arquitectura de transporte. Neste trabalho implementam-se algumas arquitecturas de QoS estudadas no IP MPLS em simuladores disponibilizados pela comunidade “Open Source” e analisamos as vantagens de desvantagens de cada uma. Todas as necessidades são devidamente equacionadas, prevendo o seu crescimento, desempenho, estabelecendo regras de distribuição de largura de banda e análise de desempenho, criando redes escaláveis e com estimativas de crescimento optimistas. Os serviços são desenhados de forma a adaptarem-se à evolução das necessidades aplicacionais, ao crescimento do número de utilizadores e evolução do próprio serviço.

Partial utility-driven scheduling for flexible SLA and pricing arbitration in clouds

Cloud SLAs compensate customers with credits when average availability drops below certain levels. This is too inflexible because consumers lose non-measurable amounts of performance being only compensated later, in next charging cycles. We propose to schedule virtual machines (VMs), driven by range-based non-linear reductions of utility, different for classes of users and across different ranges of resource allocations: partial utility. This customer-defined metric, allows providers transferring resources between VMs in meaningful and economically efficient ways. We define a comprehensive cost model incorporating partial utility given by clients to a certain level of degradation, when VMs are allocated in overcommitted environments (Public, Private, Community Clouds). CloudSim was extended to support our scheduling model. Several simulation scenarios with synthetic and real workloads are presented, using datacenters with different dimensions regarding the number of servers and computational capacity. We show the partial utility-driven driven scheduling allows more VMs to be allocated. It brings benefits to providers, regarding revenue and resource utilization, allowing for more revenue per resource allocated and scaling well with the size of datacenters when comparing with an utility-oblivious redistribution of resources. Regarding clients, their workloads’ execution time is also improved, by incorporating an SLA-based redistribution of their VM’s computational power.

Estimating Effective Slowdown of Tasks in Energy-Aware Clouds

Consolidation consists in scheduling multiple virtual machines onto fewer servers in order to improve resource utilization and to reduce operational costs due to power consumption. However, virtualization technologies do not offer performance isolation, causing applications’ slowdown. In this work, we propose a performance enforcing mechanism, composed of a slowdown estimator, and a interference- and power-aware scheduling algorithm. The slowdown estimator determines, based on noisy slowdown data samples obtained from state-of-the-art slowdown meters, if tasks will complete within their deadlines, invoking the scheduling algorithm if needed. When invoked, the scheduling algorithm builds performance and power aware virtual clusters to successfully execute the tasks. We conduct simulations injecting synthetic jobs which characteristics follow the last version of the Google Cloud tracelogs. The results indicate that our strategy can be efficiently integrated with state-of-the-art slowdown meters to fulfil contracted SLAs in real-world environments, while reducing operational costs in about 12%.

Metamodelação de processos e serviços

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para a obtenção do Grau de Mestre em Engenharia Informática

PIASA: A power and interference aware resource management strategy for heterogeneous workloads in cloud data centers

Cloud data centers have been progressively adopted in different scenarios, as reflected in the execution of heterogeneous applications with diverse workloads and diverse quality of service (QoS) requirements. Virtual machine (VM) technology eases resource management in physical servers and helps cloud providers achieve goals such as optimization of energy consumption. However, the performance of an application running inside a VM is not guaranteed due to the interference among co-hosted workloads sharing the same physical resources. Moreover, the different types of co-hosted applications with diverse QoS requirements as well as the dynamic behavior of the cloud makes efficient provisioning of resources even more difficult and a challenging problem in cloud data centers. In this paper, we address the problem of resource allocation within a data center that runs different types of application workloads, particularly CPU- and network-intensive applications. To address these challenges, we propose an interference- and power-aware management mechanism that combines a performance deviation estimator and a scheduling algorithm to guide the resource allocation in virtualized environments. We conduct simulations by injecting synthetic workloads whose characteristics follow the last version of the Google Cloud tracelogs. The results indicate that our performance-enforcing strategy is able to fulfill contracted SLAs of real-world environments while reducing energy costs by as much as 21%.

Achieving better service level results, through improvements in procedures

This project aims to prepare Worten Empresas (WE) fulfilling the increasing market demand through process changings, focusing on the Portuguese market, particularly on internal B2B clients1. Several methods were used to measure the current service level provided - process mapping, resources assessment, benchmark and a survey. The results were then used to compare against service level actually desired by WE’s customer, and then to identify the performance gaps in response times and quality of the follow-up during the sales process. To bridge the identified gaps, both a set of recommendations and an implementation plan were suggested to improve and monitor customer experience. This study concluded that it is possible to fulfill the increasing level of demand and at the same time improve customer satisfaction by implementing changes at the operations level.

A trust model for cloud computing environment

This paper presents a proposal for a management model based on reliability requirements concerning Cloud Computing (CC). The proposal was based on a literature review focused on the problems, challenges and underway studies related to the safety and reliability of Information Systems (IS) in this technological environment. This literature review examined the existing obstacles and challenges from the point of view of respected authors on the subject. The main issues are addressed and structured as a model, called "Trust Model for Cloud Computing environment". This is a proactive proposal that purposes to organize and discuss management solutions for the CC environment, aiming improved reliability of the IS applications operation, for both providers and their customers. On the other hand and central to trust, one of the CC challenges is the development of models for mutual audit management agreements, so that a formal relationship can be established involving the relevant legal responsibilities. To establish and control the appropriate contractual requirements, it is necessary to adopt technologies that can collect the data needed to inform risk decisions, such as access usage, security controls, location and other references related to the use of the service. In this process, the cloud service providers and consumers themselves must have metrics and controls to support cloud-use management in compliance with the SLAs agreed between the parties. The organization of these studies and its dissemination in the market as a conceptual model that is able to establish parameters to regulate a reliable relation between provider and user of IT services in CC environment is an interesting instrument to guide providers, developers and users in order to provide services and secure and reliable applications.

Palvelutasosopimus toimittajan hallinnnan työkaluna Case: IT-käyttötukipalvelu

Työn tavoitteena on selvittää palvelutasosopimuksella ja sen hallinnalla saatavaa hyötyä toimittajan hallinnassa. Tutkimusongelmaa on lähdetty selvittämään tutustumalla ensin lyhyesti toimittajan hallintaan ja laajemmin palvelutasosopimuksista ja niiden hallinnasta olemassa olevaan teoria-aineistoon. Tämän teoriapohjan perusteella palvelutasosopimuksia ja niiden hallintaa on tutkittu käytännössä case-yrityksen valossa, mahdollisimman realistiseen lopputulokseen pääsemiseksi. Palvelujen ulkoistaminen tulee tulevaisuudessa lisääntymään ja tätä kautta yritysten onnistuminen markkinoilla on yhä riippuvaisempaa niiden kyvystä hallita toimittajiaan. Palvelutasosopimukset ja prosessit niiden ympärillä ovat hyödyllisiä työkaluja toimittajan suorituksen valvonnassa ja kehittämisessä sekä toimittajasuhteen luotsaamisessa kohti kumppanuussuhdetta. Tämä edellyttää kuitenkin, että kauppakumppanit sisäistävät palvelutasosopimusten kautta saavutettavat hyödyt ja sitoutuvat koko palvelutason hallintaprosessiin.

Green Service Level Agreement under Sustainability Lens in IT Industry

Nowadays, when most of the business are moving forward to sustainability by providing or getting different services from different vendors, Service Level Agreement (SLA) becomes very important for both the business providers/vendors and as well as for users/customers. There are many ways to inform users/customers about various services with its inherent execution functionalities and even non-functional/Quality of Services (QoS) aspects through negotiating, evaluating or monitoring SLAs. However, these traditional SLA actually do not cover eco-efficient green issues or IT ethics issues for sustainability. That is why green SLA (GSLA) should come into play. GSLA is a formal agreement incorporating all the traditional commitments as well as green issues and ethics issues in IT business sectors. GSLA research would survey on different traditional SLA parameters for various services like as network, compute, storage and multimedia in IT business areas. At the same time, this survey could focus on finding the gaps and incorporation of these traditional SLA parameters with green issues for all these mentioned services. This research is mainly points on integration of green parameters in existing SLAs, defining GSLA with new green performance indicators and their measurable units. Finally, a GSLA template could define compiling all the green indicators such as recycling, radio-wave, toxic material usage, obsolescence indication, ICT product life cycles, energy cost etc for sustainable development. Moreover, people’s interaction and IT ethics issues such as security and privacy, user satisfaction, intellectual property right, user reliability, confidentiality etc could also need to add for proposing a new GSLA. However, integration of new and existing performance indicators in the proposed GSLA for sustainable development could be difficult for ICT engineers. Therefore, this research also discovers the management complexity of proposed green SLA through designing a general informational model and analyses of all the relationships, dependencies and effects between various newly identified services under sustainability pillars. However, sustainability could only be achieved through proper implementation of newly proposed GSLA, which largely depends on monitoring the performance of the green indicators. Therefore, this research focuses on monitoring and evaluating phase of GSLA indicators through the interactions with traditional basic SLA indicators, which would help to achieve proper implementation of future GSLA. Finally, this newly proposed GSLA informational model and monitoring aspects could definitely help different service providers/vendors to design their future business strategy in this new transitional sustainable society.

Design and Development of a Performance Evaluation Prototype for IT Organizations in the Context of Outsourcing

The present study shows design and development of a performance evaluation prototype for IT organizations in the context of outsourcing. The main objective of this research is to help an IT organization in the context of outsourcing to realize its current standing, so it can take corrective steps where ever necessary and strive for continuous improvement. Service level management (SLM) process plays a crucial role in controlling the quality provision for IT service. Out sourcing is the process of entrusting the responsibility of providing certain goods and services to an external party. We have tried to identify as many as twenty complexities and categorized in to four headings. Complexities associated with contracts and SLAs,SLM process,SLM organization and complexities due to intrinsic characteristics. In this study it is possible to measure the quality of the performance of an IT organization in an outsourcing environment effectively

Modellbasiertes Service Level Management verteilter Anwendungssysteme

Zur Senkung von Kosten werden in vielen Unternehmen Dienstleistungen, die nicht zur Kernkompetenz gehören, an externe Dienstleister ausgelagert. Dieser Prozess wird auch als Outsourcing bezeichnet. Die dadurch entstehenden Abhängigkeiten zu den externen Dienstleistern werden mit Hilfe von Service Level Agreements (SLAs) vertraglich geregelt. Die Aufgabe des Service Level Managements (SLM) ist es, die Einhaltung der vertraglich fixierten Dienstgüteparameter zu überwachen bzw. sicherzustellen. Für eine automatische Bearbeitung ist daher eine formale Spezifikation von SLAs notwendig. Da der Markt eine Vielzahl von unterschiedlichen SLM-Werkzeugen hervorgebracht hat, entstehen in der Praxis Probleme durch proprietäre SLA-Formate und fehlende Spezifikationsmethoden. Daraus resultiert eine Werkzeugabhängigkeit und eine limitierte Wiederverwendbarkeit bereits spezifizierter SLAs. In der vorliegenden Arbeit wird ein Ansatz für ein plattformunabhängiges Service Level Management entwickelt. Ziel ist eine Vereinheitlichung der Modellierung, so dass unterschiedliche Managementansätze integriert und eine Trennung zwischen Problem- und Technologiedomäne erreicht wird. Zudem wird durch die Plattformunabhängigkeit eine hohe zeitliche Stabilität erstellter Modelle erreicht. Weiteres Ziel der Arbeit ist, die Wiederverwendbarkeit modellierter SLAs zu gewährleisten und eine prozessorientierte Modellierungsmethodik bereitzustellen. Eine automatisierte Etablierung modellierter SLAs ist für eine praktische Nutzung von entscheidender Relevanz. Zur Erreichung dieser Ziele werden die Prinzipien der Model Driven Architecture (MDA) auf die Problemdomäne des Service Level Managements angewandt. Zentrale Idee der Arbeit ist die Definition von SLA-Mustern, die konfigurationsunabhängige Abstraktionen von Service Level Agreements darstellen. Diese SLA-Muster entsprechen dem Plattformunabhängigen Modell (PIM) der MDA. Durch eine geeignete Modelltransformation wird aus einem SLA-Muster eine SLA-Instanz generiert, die alle notwendigen Konfigurationsinformationen beinhaltet und bereits im Format der Zielplattform vorliegt. Eine SLA-Instanz entspricht damit dem Plattformspezifischen Modell (PSM) der MDA. Die Etablierung der SLA-Instanzen und die daraus resultierende Konfiguration des Managementsystems entspricht dem Plattformspezifischen Code (PSC) der MDA. Nach diesem Schritt ist das Managementsystem in der Lage, die im SLA vereinbarten Dienstgüteparameter eigenständig zu überwachen. Im Rahmen der Arbeit wurde eine UML-Erweiterung definiert, die eine Modellierung von SLA-Mustern mit Hilfe eines UML-Werkzeugs ermöglicht. Hierbei kann die Modellierung rein graphisch als auch unter Einbeziehung der Object Constraint Language (OCL) erfolgen. Für die praktische Realisierung des Ansatzes wurde eine Managementarchitektur entwickelt, die im Rahmen eines Prototypen realisiert wurde. Der Gesamtansatz wurde anhand einer Fallstudie evaluiert.

The assessment of benchmarks executed on bare-metal and using para-virtualization

A full assessment of para-­virtualization is important, because without knowledge about the various overheads, users can not understand whether using virtualization is a good idea or not. In this paper we are very interested in assessing the overheads of running various benchmarks on bare-­‐metal, as well as on para-­‐virtualization. The idea is to see what the overheads of para-­‐ virtualization are, as well as looking at the overheads of turning on monitoring and logging. The knowledge from assessing various benchmarks on these different systems will help a range of users understand the use of virtualization systems. In this paper we assess the overheads of using Xen, VMware, KVM and Citrix, see Table 1. These different virtualization systems are used extensively by cloud-­‐users. We are using various Netlib1 benchmarks, which have been developed by the University of Tennessee at Knoxville (UTK), and Oak Ridge National Laboratory (ORNL). In order to assess these virtualization systems, we run the benchmarks on bare-­‐metal, then on the para-­‐virtualization, and finally we turn on monitoring and logging. The later is important as users are interested in Service Level Agreements (SLAs) used by the Cloud providers, and the use of logging is a means of assessing the services bought and used from commercial providers. In this paper we assess the virtualization systems on three different systems. We use the Thamesblue supercomputer, the Hactar cluster and IBM JS20 blade server (see Table 2), which are all servers available at the University of Reading. A functional virtualization system is multi-­‐layered and is driven by the privileged components. Virtualization systems can host multiple guest operating systems, which run on its own domain, and the system schedules virtual CPUs and memory within each Virtual Machines (VM) to make the best use of the available resources. The guest-­‐operating system schedules each application accordingly. You can deploy virtualization as full virtualization or para-­‐virtualization. Full virtualization provides a total abstraction of the underlying physical system and creates a new virtual system, where the guest operating systems can run. No modifications are needed in the guest OS or application, e.g. the guest OS or application is not aware of the virtualized environment and runs normally. Para-­‐virualization requires user modification of the guest operating systems, which runs on the virtual machines, e.g. these guest operating systems are aware that they are running on a virtual machine, and provide near-­‐native performance. You can deploy both para-­‐virtualization and full virtualization across various virtualized systems. Para-­‐virtualization is an OS-­‐assisted virtualization; where some modifications are made in the guest operating system to enable better performance. In this kind of virtualization, the guest operating system is aware of the fact that it is running on the virtualized hardware and not on the bare hardware. In para-­‐virtualization, the device drivers in the guest operating system coordinate the device drivers of host operating system and reduce the performance overheads. The use of para-­‐virtualization [0] is intended to avoid the bottleneck associated with slow hardware interrupts that exist when full virtualization is employed. It has revealed [0] that para-­‐ virtualization does not impose significant performance overhead in high performance computing, and this in turn this has implications for the use of cloud computing for hosting HPC applications. The “apparent” improvement in virtualization has led us to formulate the hypothesis that certain classes of HPC applications should be able to execute in a cloud environment, with minimal performance degradation. In order to support this hypothesis, first it is necessary to define exactly what is meant by a “class” of application, and secondly it will be necessary to observe application performance, both within a virtual machine and when executing on bare hardware. A further potential complication is associated with the need for Cloud service providers to support Service Level Agreements (SLA), so that system utilisation can be audited.

Análise de contratos de tecnologia da informação (TI)

Organizações de TI deparam-se com problemas em contratos de TI (produtos e serviços) tais como: definição inadequada ou insuficiente dos produtos/serviços contratados; indefinição de responsabilidades; SLAs (Service Level Agreements) inadequados ou inexistentes; ausência ou inadequação de cláusulas de penalidades e métodos claros e objetivos de mensuração dos produtos/serviços entregues; sobreposição ou lacunas entre contratos de diferentes fornecedores

Análise histológica, histométrica e fractal de enxerto ósseo autógeno, em elevação de seio maxilar: estudo clínico

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

O SLA funk de Fernanda Abreu

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)