Academic literature on the topic 'Apache OpenWhisk'

Serverless computing is a paradigm where programmers write and compose stateless functions, leaving to Serverless infrastructure providers the duty to manage their deployment and scaling. Although the convenience of being relieved from handling the infrastructure and scaling is very appealing, there can be situations where it becomes a trade-off with efficiency. In this work we discuss several factors that current offerings don't take into account, but have an impact on efficiency and efficacy. When an inbound request to execute a function reaches the infrastructure of a provider, the latter needs to decide on which of its available computation nodes it should schedule the function. One way to improve serverless scheduling is through data locality. Intuitively, functions that access and elaborate data are faster to execute when scheduled on nodes near the data. Furthermore developers can have more control on the single nodes with self managed clusters, and can apply further optimizations for the different kind of resources to improve performance. In this work we present an extension of OpenWhisk, an open source project for serverless computing. It allows a more fine-grained control over function's scheduling. This extension provides developers with a declarative language they can use to specify priority policies for their functions. The scheduler uses those policies to find the node that best fits the constraints specified by the developer for a given function. We demonstrate the impact of data locality in a hybrid cluster context, where nodes were part of different local networks far from each other and we describe an use-case combining IoT, Edge and Cloud Computing architectures showing that we can capture the needs of different execution platforms. It is also useful to demonstrate that in such cases it is not needed to have specifically-designed platforms to host Serverless functions.

Os recentes avanços nas tecnologias de virtualização e computação em nuvem levaram o surgimento da computação sem servidor, uma tecnologia também conhecida como Function­as­a­Service. É um modelo de computação em nuvem que visa abstrair a gestão de servidores e as decisões de infraestrutura de baixo nível dos utilizadores, sendo que o utilizador cria, desenvolve e implanta funções e aplicações e a gestão do servidor fica a cargo de provedor de serviço de nuvem. Atualmente, os principais provedores de serviços em nuvem pública oferecem plataformas de computação sem servidor. No entanto, tais plataformas requerem que as funções sejam escritas ou implantadas de uma determinada maneira, o que resulta em vendor lock­in (dependência do fornecedor). Várias plataformas sem servidor open source foram propostas para permitir a execução de computação sem servidor em infraestruturas privadas, de maneira a evitar, assim, qualquer forma de dependência de fornecedores. Esta dissertação pretende comparar o desempenho da plataforma sem servidor open source Apache OpenWhisk numa implementação local, e a plataforma sem servidor da IBM Cloud, denominado de IBM Cloud Functions que usa Apache OpenWhisk como gestão para funções como serviço. No ambiente local, Apache Openwhisk foi implementada no cluster do Kubernetes, as configurações, execuções e invocações de funções foi feita usando a ferramenta de linha de comando da Apache OpenWhisk, conhecido como CLI OpenWhisk (wsk). Na IBM Cloud, as mesmas foram feitas na interface de utilizador baseada em Web da IBM Cloud Functions. Os experimentos envolveram tempo de respostas de invocações de funções usando a linguagens PHP. Para avaliação de tempo de respostas foi usada a ferramenta de monitorização Prometheus e Grafana no ambiente local, na IBM foi através IBM Cloud Functions Dashboad.<br>Recent advances in virtualization and cloud computing technologies have led to the emergence of serverless computing, a technology also known as Function­as­a­Service, a cloud computing model that aims to abstract server management and low­level infrastructure decisions away from developers, where the user creates, develops and deploys functions and applications and server management is the responsibility of the cloud service provider. Today, leading public cloud service providers offer serverless computing platforms. However, such platforms require functions to be written or implemented in a certain way, which results in a vendor lock­in (vendor dependency). Several open source serverless platforms have been proposed to allow the execution of serverless computing in private infrastructures, thus avoiding any form of dependence on suppliers. This dissertation aims to compare the performance of the open source Apache OpenWhisk serverless platform in a local implementation, and the IBM Cloud serverless platform, called IBM Cloud Functions which uses Apache OpenWhisk as management for functions as a service. In the local environment, Apache Openwhisk was implemented in the Kubernetes cluster, the configurations, executions and function invocations were done using the Apache OpenWhisk command line tool, known as CLI OpenWhisk (wsk). In the IBM Cloud, they were done in the IBM Cloud Functions web­based user interface. The experiments involved function invocation response times using the PHP languages. To evaluate the response time, the monitoring tool Prometheus and Grafana was used in the local environment, at IBM it was through IBM Cloud Functions Dashboad.<br>Este trabalho foi financiado pela FCT/MCTES através de fundos nacionais e quando aplicável cofinanciado por fundos comunitários no âmbito do projeto UIDB/50008/2020 e foi suportado pela operação Centro­01­0145­FEDER­000019 ­ C4 ­ Centro de Competências em Cloud Computing, cofinanciada pelo Fundo Europeu de Desenvolvimento Regional (FEDER) através do Programa Operacional Regional do Centro (Centro 2020), no âmbito do Sistema de Apoio à Investigação Científica e Tecnológica ­ Programas Integrados de ICDT.