Revisões de Literatura

Estou pagando uma disciplina de Análise de Desempenho e acabei de começar umas revisões de literatura. Os artigos selecionados foram:


Performance Analysis of Cloud Computing Services for Many-Tasks Scientific Computing

Alexandru Iosup, Simon Ostermann, M. Nezih Yigitbasi, Radu Prodan, Thomas Fahringer, and Dick H.J. Epema

Abstract—Cloud computing is an emerging commercial infrastructure paradigm that promises to eliminate the need for maintaining
expensive computing facilities by companies and institutes alike. Through the use of virtualization and resource time sharing, clouds
serve with a single set of physical resources a large user base with different needs. Thus, clouds have the potential to provide to their
owners the benefits of an economy of scale and, at the same time, become an alternative for scientists to clusters, grids, and parallel
production environments. However, the current commercial clouds have been built to support web and small database workloads,
which are very different from typical scientific computing workloads. Moreover, the use of virtualization and resource time sharing may
introduce significant performance penalties for the demanding scientific computing workloads. In this work, we analyze the
performance of cloud computing services for scientific computing workloads. We quantify the presence in real scientific computing
workloads of Many-Task Computing (MTC) users, that is, of users who employ loosely coupled applications comprising many tasks to
achieve their scientific goals. Then, we perform an empirical evaluation of the performance of four commercial cloud computing
services including Amazon EC2, which is currently the largest commercial cloud. Last, we compare through trace-based simulation the
performance characteristics and cost models of clouds and other scientific computing platforms, for general and MTC-based scientific
computing workloads. Our results indicate that the current clouds need an order of magnitude in performance improvement to be
useful to the scientific community, and show which improvements should be considered first to address this discrepancy between offer
and demand.
Index Terms—Distributed systems, distributed applications, performance evaluation, metrics/measurement, performance measures


The Impact of Virtualization on Network Performance of Amazon EC2 Data Center
Guohui Wang T. S. Eugene Ng

Abstract —Cloud computing services allow users to lease computing resources from large scale data centers operated by service
providers. Using cloud services, users can deploy a wide variety of applications dynamically and on-demand. Most cloud service
providers use machine virtualization to provide flexible and cost effective resource sharing. However, few studies have investigated
the impact of machine virtualization in the cloud on networking
performance. In this paper, we present a measurement study to characterize the impact of virtualization on the networking performance of the
Amazon Elastic Cloud Computing (EC2) data center. We measure the processor sharing, packet delay, TCP/UDP throughput and
packet loss among Amazon EC2 virtual machines. Our results show that even though the data center network is lightly utilized,
virtualization can still cause significant throughput instability and abnormal delay variations. We discuss the implications of our
findings on several classes of applications.

Index Terms —Measurement, cloud service, virtualization, networking performance

Weka 3: Data Mining Software in Java

Weka is a collection of machine learning algorithms for data mining tasks. The algorithms can either be applied directly to a dataset or called from your own Java code. Weka contains tools for data pre-processing, classification, regression, clustering, association rules, and visualization. It is also well-suited for developing new machine learning schemes.


http://www.cs.waikato.ac.nz/ml/weka/

Maximizando a tela do UBUNTU no VirtualBox

Um problema que sempre tenho é como maximizar a tela do UBUNTU quando utilizando o VirtualBox (pra ficar com a tela toda preenchida, sem os ícones lá de cima).

Pesquisando pela internet achei isso, que resolveu o meu problema:


1 ) No Virtual Box, quando estiver com uma máquina virtual iniciada, clica-se no menu "Dispositivos", no item "Instalar adicionais de convidados".
2 ) Já no UBUNTU, abra um terminal e digite: /media/VBOXADDITIONS_4.1.2_73507, ou algo parecido em /media.
3 ) sudo ./VBoxLinuxAdditions-x86.run

MINIPLOP 2011 - Requisitos de Qualidade: Um Padrão para Identificação de Requisitos Não Funcionais Conflitantes

Em 24/09/2011 (domingo mesmo!) estarei em São Paulo no MINIPLOP 2011 (http://www.miniplop.ita.br/TheIntroduction.htm) apresentando o artigo "Requisitos de Qualidade: Um Padrão para Identificação de Requisitos Não Funcionais Conflitantes".


Resumo. O desenvolvimento de software é uma atividade que trabalha diversos aspectos da Engenharia de Software. Um deles é a elicitação de requisitos e sua escrita em uma notação de fácil entendimento para os desenvolvedores de software. Requisitos Não Funcionais são aqueles que estão associados aos aspectos qualitativos de um software, como desempenho, segurança e usabilidade. Normalmente a identificação dos requisitos não funcionais é realizada de maneira ad hoc, e desenvolvidas tardiamente. Neste artigo é proposto um padrão para a identificação e seleção de requisitos não funcionais para aplicações comerciais nos estágios iniciais de seu desenvolvimento, que não estejam conflitantes entre si, e que estejam mais aderentes às necessidades dos clientes.

Palavras-chave: Requisitos Não Funcionais, Padrões de Software, Qualidade de Software, Engenharia de Requisitos.

Provedores de IaaS

O site http://cloud-computing.findthebest.com/saved_search/Best-IaaS-Cloud-Computing-Providers possui uma lista com vários provedores de IaaS.
Além disso, ele permite que você realize uma consulta dinâmica com vários parâmetros, como catgeoria, licenciamento, tipo de subscrição, interface de controle, suporte, sistema operacional e precificação, e comparar provedores:

Função Freqüência no MS Excel

Estava apanhando no Excel para fazer a contagem da frequencia dos elementos em um conjunto de dados, quando encontrei o site do Professor Paul Fisher, da UFRGS - http://chasqueweb.ufrgs.br/~paul.fisher/apostilas/ms_excel/func_freq/

Nele ele faz um passo a passo de como calcular a freqüência dos dados, de uma maneira simples:
Copiei o trecho do site dele (citado acima) que ensina a calcular com as figuras.
Parabéns pelo passo a passo!!!

Passo 1 - Vamos calcular a freqüência dos dados (número de ocorrências de cada valor distinta) que se encontram na figura a direta. O rotulo se encontra na célula A1 e os dados nas células A1:A10. Podemos ver que há 1 ocorrência do valor 1, 2 ocorrências do valor 2, 3 ocorrências do valor 3, 2 ocorrências do valor 4 e 1 ocorrência do valor 5. O que nos queremos, ao final, é uma tablela:

valor	freqüência
1	1
2	2
3	3
4	2
5	1

---

Passo 2 - A função freqüência precisa de dois variaveis de entrada: uma matiz de dados e uma matriz de bins. A matriz de bins serve para indicar as categorias a ser usadas na execução da função. Então, precisamos criar uma matriz de categorias que aparece na figura a direta. O conteúdo da célula C2 servirá para contar as ocorrências dos valores <=1 (menor que ou igual a 1). O conteúdo da célula C3 servirá para contar as ocorrências dos valores <=2 (menor que ou igual a 2) e >1 (maior que 1). O conteúdo da célula C6 servirá para contar as ocorrências dos valores <=6 e >5.

---

Passo 3 - A função "FREQÜÊNCIA" é uma função matriz. Isso significa que o destino do resultado da execução da função precisa ser uma matiriz de células e não uma célula só. Precisamos selecionar a matriz de células em qual será colocado os resulatdos antes "chamar" a função freqüência. Sendo que nossos categorias se encontram em C2:C6, digitaremos o rótulo "freqüência" na célula D1 e selcionaremos as células D2:D6 para receber os resultados.

---

Passo 4 - Com as células D2:D6 selecionadas, clicaremos em "inserir" no menu principal e "função" no sub-menu. Do diálogo que aparece selcionaremos a função "freqüência" se encontra na categoria "estatística" e, finalmente em OK.

---

Passo 5 - Um diálago aparece pedindo preenchimento da matriz dos dados e a matriz dos bins. Isso pode ser feito digitando diretamente nos respetivos campos OU, com o cursor no respetivo campo, selecionando a extensão de células com os dados e os bins usando o mouse.

IMPORTANTE! Não clique em OK

---

Passo 6 - Em vez de cliquar em OK, manteremos apertados as botões "Ctrl" e "Shift" enquanto clicamos o botõe "Enter". As freqüências aparecem ao lado das categorias, neste caso, ou em qualquer conjunto de células (matriz) inicialmente selcionado para receber o resultado.Porque ter um botão OK se não serve para executar a função? Não sei ... bobagem da Microsoft. Todas as funções matriz em Excel funcionam assim.

Fonte: site do Professor Paul Fisher, da UFRGS.
(http://chasqueweb.ufrgs.br/~paul.fisher/apostilas/ms_excel/func_freq/)

Cloud Computing and Scientific Applications (CCSA 2011)

1st International Workshop on
Cloud Computing and Scientific Applications (CCSA 2011)
http://www.cloudbus.org/ucc2011/ccsa/ccsa2011.html

In conjunction with the 4th IEEE International
Conference on Utility and Cloud Computing (UCC 2011) December 5-7,
Melbourne, Australia http://www.cloudbus.org/ucc2011

***IBM R&D Australia is sponsoring "Best Student Paper Award"
among the papers accepted for presentation at the workshop***
--Selection and final decision resides at the PC members discretion

CCSA workshop has been formed to promote research and development
activities focused on enabling and scaling scientific applications using
distributed computing paradigms, such as cluster, Grid, and Cloud
Computing. With the rapid emergence of software systems and their
applicability, the volume of users are growing exponentially. User
requirements are getting more and more complex. Existing computing
infrastructure, software system designs, and use cases will have to take
into account the enormity in volume of requests, size of data, computing
load, locality and type of users, and so forth.

Cloud computing promises reliable services delivered through
next-generation data centers that are built on compute and storage
virtualization technologies. Users will be able to access applications
and data from a ÒCloudÓ anywhere in the world on demand. In other words,
the Cloud appears to be a single point of access for all the computing
needs of users. The users are assured that the Cloud infrastructure is
robust and will always be available at any time.

CCSA brings together researchers and practitioners from around the world
to share their experiences on modeling, executing, and monitoring
scientific applications on Clouds.

In this workshop, we are interested in receiving innovative work on
enabling and scaling computing systems to support the execution of
scientific applications. The target audience include researchers and
industry practitioners who are interested in distributed systems,
particularly focusing on scaling of applications using Cloud computing.



TOPICS OF INTEREST
-------------------

Topics of interest include (but are not limited to):

- Enabling applications using distributed systems
- Architectural Models for scaling of applications
- Novel applications for cloud computing, including games and social networks
- Novel cloud programming models
- Innovative cloud service models
- Support for scalable and elastic cloud services
- Cloud support for mobile applications and Content Delivery Networks
- Reliability of applications and services running on the cloud
- Performance monitoring for cloud applications
- Cloud use case studies
- Scientific computing in the cloud
- Business computing in the cloud
- Social computing in the cloud


IMPORTANT DATES
---------------

Paper Submission Deadline: September 13, 2011 (Extended due to Author Requests)
Author Notification: September 25, 2011
Camera Ready Submission: September 30, 2011


PROGRAM CHAIRS
---------------

Suraj Pandey
The Commonwealth Scientific and Industrial Research Organisation (CSIRO)

Andrew Melatos
School of Physics, The University of Melbourne


INTERNATIONAL PROGRAM COMMITTEE
--------------------------------
Albert Zomaya, University of Sydney, Australia
Alberto Sanchez, Universidad Rey Juan Carlos, Spain
Craig Lee, Open Grid Forum, USA
Pavan Balaji, Argonne National Laboratory, USA
Ivona Brandic, Vienna University of Technology, Austria
Klaus-Dieter Schewe, Software Competence Center, Austria
Bahman Javadi, University of Melbourne, Australia
Bart Pindor, University of Melbourne, Australia
Richard O. Sinnott, e-Research at University of Melbourne, Australia
James Sankar, AARNet Pty Ltd, Australia
Christian Simone Vecchiola, IBM Research and Development Australia
Surya Nepal, CSIRO, Australia
Bruno Schulze, National Laboratory for Sci. Computing, Brazil
Manish Parashar, Rutgers, The State University of New Jersey, USA


PAPER SUBMISSION GUIDELINES
----------------------------

All papers must be submitted electronically and in PDF format through easychair.

https://www.easychair.org/conferences/?conf=ccsa2011

The material presented should be original and not published or under
submission elsewhere. Authors should submit full papers of up to 6
pages, strictly following the IEEE Computer Society Proceedings
Manuscript style (available at
http://www.computer.org/portal/web/cscps/formatting),
using two-column, single-space format, with 10-point font size.
Figures and references must be included in the 6 pages. Oversized
papers will be automatically rejected by the committee. At least
one of the authors of each accepted paper must register early to
attend the conference, in order for the paper to appear in the
conference proceedings.

Submitted papers must represent original unpublished research
that is not currently under review for any other conference
or journal. Papers not following these guidelines will be
rejected without review and further action may be taken,
including (but not limited to) notifications sent to the
heads of the institutions of the authors and sponsors of the
conference. Submissions received after the due date,
exceeding length limit, or not appropriately structured may
also not be considered. The proceedings will be published and
will be made online through the IEEE Xplore.

JOURNAL SPECIAL ISSUE
---------------------
All papers, after presentation at the conference, will be invited to
submit their revised version to a special issue of the IJCSE -
International Journal of Computational Science and Engineering
(published by Inderscience). IJCSE is a refereed international journal
providing an international forum to report, discuss and exchange
experimental results, novel designs, work-in-progress, experience, case
studies, and trend-setting ideas in the area of computational science
and engineering. It is abstracted and indexed in ACM Guide to Computing
Literature, Computer and Information Systems, Computer Database, EI
Compendex, Engineered Materials Abstracts, Google Scholar, Pascal,
Scirus , and Scopus. For additional information about IJCSE, please
visit

http://www.inderscience.com/browse/index.php?journalCODE=ijcse