Channels ▼
RSS

Web Development

Cloud Computing Research Using Google-IBM CluE Cluster



Researchers from the San Diego Supercomputer Center (SDSC) have been awarded a two-year, $450,000 grant from the National Science Foundation (NSF) to explore new ways for academic researchers to manage computing clouds -- extremely large data sets hosted on massive, Internet-based commercial computer clusters.

The NSF award focuses on the Cluster Exploratory (CluE), a distributed, large-scale computing resource formed in late 2007 between Google and IBM. The NSF joined the Google-IBM partnership early last year, hailing the CluE initiative as a partnership between private enterprise and the federal science agency to expand access to this research infrastructure to academic institutions across the nation. Last April, the NSF issued a solicitation for research projects aimed at developing software to make CluE a researcher-friendly resource to analyze and manage extremely large amounts of data.

Specifically, SDSC researchers will explore the use of compute clouds to dynamically provision and manage large-scale scientific datasets. This is in contrast to the current approach using a traditional parallel relational database management system (RDBMS) architecture, which is more structured but also more static. The SDSC team will investigate the feasibility of the cloud computing approach versus known conventional approaches, while evaluating the trade-offs, advantages, and disadvantages.

"The CluE system provides access to a cloud computing environment characterized by relatively vast amounts of computational and storage resources," said SDSC Distinguished Scientist Chaitan Baru, who is heading the SDSC research project, called 'Performance Evaluation of On-Demand Provisioning of Data-Intensive Applications.' "This creates opportunities to rethink some of our strategies and ask ourselves some key questions: Could we use more dynamic strategies for resource allocation? Can this result in better overall performance for the user?"

Cloud computing is seen as a possible way for researchers to move from processing and managing their own data sets locally, to relying on large, off-site, commercially managed data clusters.

Amazon.com, for example, although primarily an e-commerce retailer, has made pay-as-you-go, on-demand computing and storage available via its "Elastic Compute Cloud" or EC2 platform. Introduced in mid-2006, EC2 is now being used by both startup companies and established businesses as a 'virtual' computing resource.

Like many other supercomputer scientists, Baru is concerned that the ever-increasing volume of scientific data is beginning to overwhelm current approaches to data management.

"The broader impact of this research will be to reassess how scientific data archives are implemented, and how data sets are hosted and served to the scientific community at large, using on-demand and dynamic approaches for provisioning data sets as opposed to the current static approach," said Baru. "This project has the potential to offer scientific researchers compute clouds as a complement to conventional supercomputing architectures used today, while creating new tools and techniques for commercial cloud computing."

SDSC researchers will use the Google-IBM CluE cluster in combination with the Apache Hadoop programming environment, an open-source volunteer project developed under the Apache Software Foundation, a not-for-profit enterprise dedicated to open, collaborative software development projects by supplying hardware, communication, and business infrastructure. During the next 24 months, researchers will evaluate hybrid approaches that dynamically deploy a database instance for some subsets of the data, based on user directives or workload analysis, while the rest of the data are served using Hadoop.

"The resource-rich CluE environment represents the type of systems that are becoming available to real-world applications, as opposed to the resource-constrained environments that are currently common in most production systems for large scientific data management, especially in academia," said Baru. "The Google-IBM CluE cluster is a natural environment for implementing GLW, which is highly data-parallel in nature. Our plan is to evaluate a series of implementations using the database system, as well as the distributed file system and MapReduce implementations provided by Hadoop."

In addition, SDSC researchers believe that the CluE-based or similar cluster-based implementations can be exposed to user communities via a Services Oriented Architecture (SOA) for access via other scientific workflow environments, visualization tools, and portals, such as the GEON Portal and GEON's Open Topography portal, which now contains GEON's LiDAR data access and processing capabilities.

"This in turn will bring the benefits of massively-scaled computing resources to a large community of users," noted Baru.


Related Reading


More Insights






Currently we allow the following HTML tags in comments:

Single tags

These tags can be used alone and don't need an ending tag.

<br> Defines a single line break

<hr> Defines a horizontal line

Matching tags

These require an ending tag - e.g. <i>italic text</i>

<a> Defines an anchor

<b> Defines bold text

<big> Defines big text

<blockquote> Defines a long quotation

<caption> Defines a table caption

<cite> Defines a citation

<code> Defines computer code text

<em> Defines emphasized text

<fieldset> Defines a border around elements in a form

<h1> This is heading 1

<h2> This is heading 2

<h3> This is heading 3

<h4> This is heading 4

<h5> This is heading 5

<h6> This is heading 6

<i> Defines italic text

<p> Defines a paragraph

<pre> Defines preformatted text

<q> Defines a short quotation

<samp> Defines sample computer code text

<small> Defines small text

<span> Defines a section in a document

<s> Defines strikethrough text

<strike> Defines strikethrough text

<strong> Defines strong text

<sub> Defines subscripted text

<sup> Defines superscripted text

<u> Defines underlined text

Dr. Dobb's encourages readers to engage in spirited, healthy debate, including taking us to task. However, Dr. Dobb's moderates all comments posted to our site, and reserves the right to modify or remove any content that it determines to be derogatory, offensive, inflammatory, vulgar, irrelevant/off-topic, racist or obvious marketing or spam. Dr. Dobb's further reserves the right to disable the profile of any commenter participating in said activities.

 
Disqus Tips To upload an avatar photo, first complete your Disqus profile. | View the list of supported HTML tags you can use to style comments. | Please read our commenting policy.
 
Dr. Dobb's TV