Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Cloud computing for science and engineering
Foster I., Gannon D., The MIT Press, Cambridge, MA, 2017. 392 pp. Type: Book (978-0-262037-24-2)
Date Reviewed: Feb 1 2018

Clouds have been around for some time and are often touted as expensive and complex systems meant to solve problems specific to Web companies with global reach. Foster and Gannon dispel this myth through an elegantly crafted tour of everyday data analytics tasks, with an emphasis on scientific applications, showing how to deploy cost-effective solutions through lots of examples with real code, while discussing the major concepts underlying this technology. This book is a lucid eye-opener to an ongoing paradigm shift in everyday computing.

Clouds are clusters of servers that cooperate in many ways to support large-scale and heterogeneous applications via Web services. A major motivation for the development of clouds was the need to repurpose the computing infrastructure for different applications, something that is very costly or impossible to do with dedicated high-performance clusters. The first successful clouds were developed by Web companies such as Amazon and Google with the help of academics. Today, such infrastructure is available on a “pay-as-you-go” model for everyone to use. The key computing abstraction that makes clouds so effective is encapsulating computational tasks as services, typically deployed through preconfigured virtual machines (VMs). This allows the cloud to adjust to the application demand by starting or shutting down VMs accordingly. A minimalistic cloud, readily available as an open-source solution, like OpenStack or Eucalyptus, offers just that: support for managing VMs. Other cloud software infrastructure exists, for example, to support distributed data management and processing infrastructure that supports high-end analytics applications. Distributed file systems such as the Hadoop distributed file system (HDFS) and parallel programming platforms like MapReduce (and its open-source incarnation Hadoop) are tried and true technologies that support many critical applications today. Another success story of clouds is that they have been used to support a wide range of end-user applications, from file and photo sharing in the early days, to entire office suites today. Clouds have transformed how we develop and disseminate software and data online, and will soon disrupt how we deploy everyday software solutions that can leverage sophisticated tools for image and language understanding deployed as services.

Cloud computing for science and engineering by Foster and Gannon covers all of the topics mentioned in the previous paragraph while making a very compelling case for the use of cloud infrastructure to solve massive problems faced by scientists and engineers alike. The book is supported by a wealth of coding examples illustrating the concepts, available through the accompanying website, that will get users started in no time. Most examples are quite advanced, covering nontrivial tools, and thus serve as invaluable starting points for enthusiasts. The book covers different aspects of clouds, with an emphasis on typical tasks performed by real users, instead of focusing on the technology for its own sake. First, the book covers various data storage and manipulation tools. Next it covers the fundamental abstractions and technologies involved in deploying and managing computation in the cloud, including a discussion of the support for high-performance computing as understood by scientists and engineers in commercial clouds. The last major part of the book discusses various machine learning toolkits offered by cloud providers and how to integrate them into other programs with popular languages like Python.

This book is more than an excellent practical guide. While the writing is application oriented, the authors do not miss an opportunity to relate the discussion to core concepts in computing science, providing many pointers to the related literature. In fact, I can envision teachers using this book in support of their courses. The book also discusses big questions related to cloud computing such as privacy, security, and the various trade-offs related to using commercial clouds versus procuring your own hardware. Finally, the book is alive and ever-evolving online, with new and revised material already available at no charge. All in all, this book is a remarkable achievement that bridges the major conceptual and practical aspects of using cloud infrastructure for high-performance computing. It is a must read.

More reviews about this item: Amazon

Reviewer:  Denilson Barbosa Review #: CR145825 (1805-0187)
Bookmark and Share
  Reviewer Selected
Featured Reviewer
 
 
Distributed Systems (C.2.4 )
 
 
Engineering (J.2 ... )
 
Would you recommend this review?
yes
no
Other reviews under "Distributed Systems": Date
The evolution of a distributed processing network
Franz L., Sen A., Rakes T. Information and Management 7(5): 263-272, 1984. Type: Article
Jul 1 1985
A geographically distributed multi-microprocessor system
Angioletti W., D’Hondt T., Tiberghien J.  Concurrent languages in distributed systems: hardware supported implementation (, Bristol, UK,871985. Type: Proceedings
Oct 1 1985
A fault tolerant LAN with integrated storage, as part of a distributed computing system
Boogaard H., Bruins T., Vree W., Reijns G.  Concurrent languages in distributed systems: hardware supported implementation (, Bristol, UK,1001985. Type: Proceedings
Aug 1 1985
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy