Building a High Performance Cluster with Amazon Web Services
Amazon Web Services has released the following video that provides a fascinating look at how straightforward it is to create, launch and monitor high performance cluster instances.
CPU utilization, disk I/O and network utilization are tracked as part of the metrics, and guidance on how to define hardware virtualization (HVM) is also defined. Creating an 8-node, 64 core, ad hoc cluster is defined in the steps in this video with the intent of running a molecular dynamics simulation.
What is interesting about this video is how Amazon Web Services continues to show the practicality of its broad spectrum of server capacities on the Elastic Compute Cloud (EC2). This is the first in a series of videos Amazon Web Services will be releasing on creating high performance clusters. It’s worth checking out as the walk-through of steps shows how rapidly EC2 is maturing as an enterprise platform.
Implications for the Enterprise
EC2 has language-agnostic Web Services APIs that show potential for integrating legacy systems, databases, master data management (MDM), CRM and enterprise systems. For enterprises that have data-centric operations and business models, EC2 could become the foundation of contextual search and role-based access of their legacy data. Decades of data accessed via contextual search would provide insights that aren’t possible today using existing methods of data access, integration and analysis.
Bottom line: Creating high performance clusters in AWS EC2 shows potential to increase the accuracy and precision of business intelligence and analytics, and potentially solve the most complex data-driven challenges of social CRM.
Flickr attribution: http://www.flickr.com/photos/vitroids/2586785504/