Cyclopic Energy provides independent consulting services related to wind power installations in complex environments, pedestrian wind studies, and pollutant dispersion modeling. The company specializes in optimizing the layout of small scale wind turbines by identifying issues that will affect energy yield performance and the turbines’ ability to integrate with the local environment. Cyclopic Energy’s three employees are located in Adelaide and Melbourne, Australia.
Cyclopic Energy began its operation using a traditional in-house infrastructure consisting of an eight-core server connected to several workstations. However, the company wanted to improve its ability to handle increased computing requirements and found Amazon Web Services (AWS) to be a convenient and cost-effective solution. Rick Morgans, Technical Director (CTO) explains: “Our usage profile means we have intense periods of heavy usage followed by long periods of no activity. To build and maintain our own cluster capable of delivering peak computational demand would be prohibitively expensive and inefficient.”Amazon Elastic Compute Cloud (Amazon EC2) allows Cyclopic Energy to run OpenFOAM® , an open- source Computational Fluid Dynamics software produced by OpenCFD Ltd, on eight cores simultaneously within one extra large instance. The OpenFOAM software simulates the environmental conditions at Cyclopic Energy’s wind power sites. The company typically runs its Amazon EC2 instance for intervals up to thirty hours, during which time it is able to run multiple simulations at the same time. Cyclopic Energy reports that its extra large instance within Amazon EC2 recently helped the company to perform two months worth of simulations in a two-day period.
Morgans says, “We chose AWS because it provided a simple path to speed up our service delivery. Like most consultancies, our clients expect results ASAP. AWS offers complete access to virtual instances of meaningful power and capacity, at a reasonable price. While there may be a performance degradation running on a virtual instance as compared to equivalent physical hardware, the ability to run multiple simulations concurrently means the total delivery time is much reduced.”
After utilizing Amazon EC2 to perform its simulations, Cyclopic Energy conducts is final analytical work on its in-house server. The company uses Amazon Simple Storage Service (Amazon S3) to store the data resulting from its analyses. Like Amazon EC2, Amazon S3 reduces Cyclopic Energy’s need for physical hardware and helps the company to perform its operations and access its data from any geographic location.
Cyclopic Energy plans to continue to use various Amazon EC2 instances for its high capacity computing needs. “AWS is constantly innovating,” says Morgans “Since we started using Amazon EC2, AWS has introduced larger virtual machines with fast interconnects (cluster compute instances), much like traditional supercomputing hardware, along with graphics processing unit (GPU)-based instances. We intend to move our calculations to these instances, as we believe they will offer the best price/performance tradeoff.” The company also plans to further automate its workflow with Amazon Simple Queue Service (Amazon SQS) and Amazon’s non-relational database, Amazon SimpleDB.
Cyclopic Energy credits AWS with helping it to increase its agility and responsiveness. Morgans says, “AWS makes it possible for us to deliver state-of-the-art technologies to clients within timeframes that allow us to be dynamic, without having to make large investments in physical hardware.”
Amazon Case Study: Cyclopic Energy
Date: 28-08-2013