StarHPC — Teaching parallel programming within elastic compute cloud

Ceraj, Ivica; Riley, Justin; Shubert, Charles

doi:10.1109/iti.2009.5196108

Cited by 47 publications

(32 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…With that in mind, [35] shows the case of a private cloud shared by four universities that allowed for the provision of virtual machines with preconfigured images created on demand by the students in a computer sciences course. In a different contribution, we were able to observe StarHPC, developed by the Massachusetts Institute of Technology (MIT), that proposes virtual machines that can be reused among students in a parallel programming course [36].…”

Section: Literature Reviewsmentioning

confidence: 99%

Model to Implement Virtual Computing Labs via Cloud Computing Services

Encalada

Sequera

2017

Symmetry

View full text Add to dashboard Cite

Abstract:In recent years, we have seen a significant number of new technological ideas appearing in literature discussing the future of education. For example, E-learning, cloud computing, social networking, virtual laboratories, virtual realities, virtual worlds, massive open online courses (MOOCs), and bring your own device (BYOD) are all new concepts of immersive and global education that have emerged in educational literature. One of the greatest challenges presented to e-learning solutions is the reproduction of the benefits of an educational institution's physical laboratory. For a university without a computing lab, to obtain hands-on IT training with software, operating systems, networks, servers, storage, and cloud computing similar to that which could be received on a university campus computing lab, it is necessary to use a combination of technological tools. Such teaching tools must promote the transmission of knowledge, encourage interaction and collaboration, and ensure students obtain valuable hands-on experience. That, in turn, allows the universities to focus more on teaching and research activities than on the implementation and configuration of complex physical systems. In this article, we present a model for implementing ecosystems which allow universities to teach practical Information Technology (IT) skills. The model utilizes what is called a "social cloud", which utilizes all cloud computing services, such as Software as a Service (SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS). Additionally, it integrates the cloud learning aspects of a MOOC and several aspects of social networking and support. Social clouds have striking benefits such as centrality, ease of use, scalability, and ubiquity, providing a superior learning environment when compared to that of a simple physical lab. The proposed model allows students to foster all the educational pillars such as learning to know, learning to be, learning to live together, and, primarily, learning to do, through hands-on IT training from a MOOCs. An aspect of the model has been verified experimentally and statistically through a course of computer operating systems.

show abstract

Section: Literature Reviewsmentioning

confidence: 99%

Model to Implement Virtual Computing Labs via Cloud Computing Services

Encalada

Sequera

2017

Symmetry

View full text Add to dashboard Cite

show abstract

“…De acordo com [Ivica et al 2009], além dessas despesas, pode-se citar ainda custos de hardware, custos associadosà hospedagem, energia e refrigeração para um cluster típico. Uma outra configuração desejável, para um laboratório,é alterar um cluster para utilização em salas de aula, onde podem ser necessários pacotes diferenciados, além da criação de contas de estudantes e manuais de acesso para professores e usuários.…”

Section: Abordagem Práticaunclassified

Mapeamento Sistemático do Ensino Teórico e Prático de Programação Paralela

Bachiega¹,

Souza²,

Bruschi³

et al. 2017

Anais Dos Workshops Do VI Congresso Brasileiro De Informática Na Educação (CBIE 2017)

View full text Add to dashboard Cite

Abstract. Providing parallel programming education is an emerging challenge, requires teaching methods to promote the learning process and also an elaborate infrastructure to provide a suitable environment for practical laboratory classes. Also, conventional teaching methods are used to share knowledge in the area, sometimes not being appropriate for this purpose. Effective learning methods, software configuration facility and the necessary laboratory infrastructure are needed. This paper presents a survey of recent research and challenges about parallel programming teaching. The main contribution of this work is to group state of the art in this context, highlighting the teaching practices available in the literature, to determine consolidated initiatives in development.Resumo. Fornecer educação em programação paralelaé um desafio emergente, requer métodos de ensino para promover o processo de aprendizagem e também uma infraestrutura para fornecer um ambiente adequado para as aulas de laboratório. Além disso, os métodos de ensino convencionais que são utilizados para compartilhar o conhecimento,às vezes não são adequados para esse propósito. Métodos de aprendizagem eficazes, configuração de software e uma infraestrutura de laboratório são requisitos necessários. Assim, este artigo apresenta uma revisão sistemática sobre o ensino de programação paralela. A principal contribuição deste trabalhoé reunir o estado da arte neste contexto, destacando as principais práticas de ensino disponíveis na literatura. IntroduçãoO ensino de Computação de Alto Desempenho ou HPC (High-Performance Computing), ao contrário de outros tipos de computação, aborda desafios e condições especiais. O ensino abrange grandes quantidades de informações, problemas computacionais e a compreensão das plataformas suportadas, onde os alunos precisam de capacidades e habilidades para operar esses ambientes especiais [Shamsi et al. 2015].[ Zarza et al. 2012] destacam a importância de HPC como uma ferramenta valiosa para a sociedade. HPC fornece o desenvolvimento de uma grande quantidade de aplicativos e serviços. Também abordam que o estudo de arquiteturas paralelasé uma das questões-chave para estudantes acadêmicos de computação, devido ao uso geral de computadores em diversas disciplinas.A Programação Paralela, uma parte de HPC, tornou-se ativa e intrínsecaàs tecnologias atualmente disponíveis, principalmente considerando os distintos modelos de programação provenientes dessas diferentes arquiteturas paralelas.

show abstract

“…The inherent flexibility and absence of capital investment costs have made cloud an attractive platform for bursty HPC workloads [24,64,58,51] and some educational HPC clusters [45,29]. The downsides are higher operational costs and virtualizatrion overheads [46,31,42].…”

Section: Related Workmentioning

confidence: 99%

Multi-objective job placement in clusters

Blagodurov

Fedorova

Vinnik

et al. 2015

Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

One of the key decisions made by both MapReduce and HPC cluster management frameworks is the placement of jobs within a cluster. To make this decision, they consider factors like resource constraints within a node or the proximity of data to a process. However, they fail to account for the degree of collocation on the cluster's nodes. A tight process placement can create contention for the intra-node shared resources, such as shared caches, memory, disk, or network bandwidth. A loose placement would create less contention, but exacerbate network delays and increase cluster-wide power consumption. Finding the best job placement is challenging, because among many possible placements, we need to find one that gives us an acceptable trade-off between performance and power consumption. We propose to tackle the problem via multi-objective optimization. Our solution is able to balance conflicting objectives specified by the user and efficiently find a suitable job placement.

show abstract

StarHPC — Teaching parallel programming within elastic compute cloud

Cited by 47 publications

References 3 publications

Model to Implement Virtual Computing Labs via Cloud Computing Services

Model to Implement Virtual Computing Labs via Cloud Computing Services

Mapeamento Sistemático do Ensino Teórico e Prático de Programação Paralela

Multi-objective job placement in clusters

Contact Info

Product

Resources

About