HPC – The New Secret Weapon!

Manisha Bhargava
Head - Innovation Practice
Wipro Systems & Technology

The custom supercomputer of yesteryear has given way to commodity-based supercomputing, or what is now called High Performance Computing (HPC). In today's HPC world, it is not uncommon for the supercomputer to use the same hardware found in Web servers and even desktop workstations. The HPC world is now open to almost everyone because the cost of entry is at an all-time low.

To many organizations, HPC is now considered an essential part of business success. They won't talk much about it because it is considered a competitive advantage. They are starting to design new products, optimizing manufacturing and delivery processes, solving production problems, mining data and simulating everything from business process to shipping crates all in an effort to become more competitive, profitable and “green”. 

HPC represents a tremendous competitive edge in the marketplace because it can give users the ability to quickly model and then manipulate a product or process to see the impact of various decisions before they are made. Consider a car as it moves through the air. Once it has been modelled in an HPC system, the end-user can look at the air flow at points that would be almost impossible with a physical model in a wind tunnel. The possibilities are endless. Existing products and processes can be optimized to reduce cost, and new avenues of development can be explored at a lower cost and with faster time to market.

Forms of HPC

Applying compute cycles to a problem can be done in many ways. Currently, there seem to be four modes in which the cycles needed can be obtained for typical HPC problems:

• The commodity HPC cluster: Over the last ten years, the HPC cluster has disrupted the entire supercomputing market. Built from standard off-the-shelf servers and high-speed interconnects, a typical HPC system can deliver industry-leading, cost-effective performance. 

• Dedicated supercomputer: In the past, the dedicated supercomputer was the only way to throw a large number of compute cycles at a problem. Supercomputers are still produced today and often use specialized non-commodity components. 

• HPC cloud computing: This method is relatively new and employs Internet as a basis for a cycles-as-a-service model of computing. The compute cycles in question live in the cloud somewhere allowing a user to request remote access to cycles on-demand. An HPC cloud provides dynamic and scalable resources (and possibly virtualization) to the end-user as a service. 

• Grid computing: Grid is similar to cloud computing, but requires more control by the end-user. Its main use is academic projects where local HPC clusters are connected and shared on a national and international level. Some computational grids span the globe, while others are located within a single organization.

Benefits of HPC

A typical HPC set-up consists of Servers, Networking, Storage, Software and Services (Assessment, Deployment, Training and Support). The benefits of High Performance Computing to organizations are immense. It helps provide solutions to problems that need significant computational power, access and process large amounts of data and operate interactively across a geographically distributed network. HPC can also be used to replace scale-up SMP supercomputers at 10-20% of the cost lower than hardware acquisition and service/support costs

Wipro and HPC

Wipro has been the leader in changing the landscape of HPC implementation in India across various industry segments, including Oil & Gas, Research, Banking, Life Sciences, and Intelligence & Academics. Coupled with the latest industry standard technology and its understanding of the HPC environment, the advantages that Wipro has to offer include:

• Industry-leading standards-based product, not proprietary hardware, through our Wipro NetPower Datacenter Systems
• Complete integration of all components in the HPC stack
• Standard and Customized Supercomputer
• Leader in power/cooling and high-density servers
• Customer and standard benchmark capabilities 
• Best-of-breed cluster components offering the best of solutions to customers
• Optimal use of investments and better RoI
• Scalable Computing from a few T Flops to several Peta Flops
• Latest technology-based CPUs and low latency, high bandwidth interconnects
• Turnkey product delivery