Amazon Launches Clusters in the Cloud

By DE Editors

Amazon Web Services has announced the availability of Cluster Compute Instances for Amazon EC2, a new instance type designed for high-performance computing (HPC) applications and other demanding network-bound applications. It is designed for customers with complex computational workloads, such as tightly coupled parallel processes, or with applications sensitive to network performance.

According to Amazon, Cluster Compute Instances have been specifically engineered to provide high-performance network capability and can be programmatically launched into clusters “allowing applications to get the low-latency network performance required for tightly coupled, node-to-node communication. Cluster Compute Instances also provide significantly increased network throughput than Amazon’s standard EC2 offerings.

The company says using Amazon EC2 to solve large computational problems allows organizations to:

• Eliminate the cost and complexity of procuring, configuring and operating in-house compute clusters.
  
• Enable quicker innovation through access to additional compute resources.
  
• Scale compute resources to the size appropriate for each workload.
  
• Avoid incurring cost when compute resources are no longer needed.
  
 

The idea of cloud computing on clusters is catching on with engineers. MathWorks now enables its customers, using MATLAB and Parallel Computing Toolbox on their desktops, to scale data-intensive computation problems up to access greater compute power with Cluster Compute Instances for Amazon EC2, which are running MATLAB Distributed Computing Server.

“Cluster Compute Instances give MATLAB users the opportunity to test and run their high performance computing problems for data-intensive applications in the cloud at a price and performance level that allows us to continually innovate and meet customer needs,” says Silvina Grad-Freilich, senior manager, Parallel-Computing at MathWorks. “Were thrilled to allow our customers to leverage Amazon Web Services as an easily accessible way to meet their needs for increased compute power.”

Each instance provides a predictable amount of dedicated compute capacity and is charged per instance-hour consumed. There is no minimum fee and Amazon provides an online calculator to estimate costs.

The Cluster Compute instance family has the following specifications:

• 23GB of memory
  
• 33.5 EC2 Compute Units (2 x Intel Xeon X5570, quad-core Nehalem architecture)
  
• 1,690GB of instance storage
  
• 64-bit platform
  
• 10 Gigabit Ethernet
 

“Businesses and researchers have long been utilizing Amazon EC2 to run highly parallel workloads ranging from genomics sequence analysis and automotive design to financial modeling. At the same time, these customers have told us that many of their largest, most complex workloads required additional network performance,” says Peter De Santis, General Manager of Amazon EC2. “For perspective, in one of our pre-production tests, an 880 server sub-cluster achieved 41.82 TFlops on a LINPACK test run.”

For more information, visit Amazon Web Services.

Sources: Press materials received from the company and additional information gleaned from the company’s website.