Overcoming the Learning Curve of New Processor Architectures

Print Friendly, PDF & Email

In this week’s Sponsored Post, Suresh Aswani, Senior Manager, Solutions Marketing, at Hewlett Packard Enterprise, shares how to overcome the learning curve of new processor architectures. 

Suresh Aswani, Senior Manager, Solutions Marketing, Hewlett Packard Enterprise

The financial services industry is arguably one of the most rapidly evolving business landscapes that exist today. In an environment easily disrupted by the health of the economy and advancements in technology, agility and the adoption of new processor architectures has become a prerequisite for survival.

The aftermath of the financial crisis has brought heightened financial regulatory pressures and an increasing focus on costs. To grow revenue, improve margins, and maintain customer loyalty, today’s financial services firms understand they must continually modernize their technology infrastructure in order to remain relevant.

High-performance computing tools are helping financial firms survive and thrive in this highly demanding and data-intensive industry. As financial models grow in complexity and greater amounts of data must be processed and analyzed on a daily basis, firms are increasingly turning to HPC solutions to exploit the latest technology performance improvements and new processor architectures.

As financial models grow in complexity and greater amounts of data must be processed and analyzed on a daily basis, firms are increasingly turning to HPC solutions to exploit the latest technology performance improvements.

Recent research by IDC found that HPC usage in the financial sector experienced a period of hyper growth over the last few years, so much that the firm had to restate their forecasts for this sector at least 50 percent higher as compared to initial estimates.

The advent of multi-core processors has forever changed the HPC landscape, and today these types of processors are the norm. The processor ecosystem now evolves at a rapid and unpredictable rate in order to address the increasing demands of HPC users. For example, the Intel Xeon Phi product line evolved from a coprocessor to an autonomous CPU in a single generation.

[clickToTweet tweet=”The advent of multi-core processors has forever changed the HPC landscape.” quote=”The advent of multi-core processors has forever changed the HPC landscape.”]

However, this rapid evolution is leaving users struggling to overcome a steep learning curve in order to fully exploit the latest hardware innovations and make computing as efficient as possible. Specifically for software developers, here is the challenge: How to quickly gain an understanding of the latest multi-core, multi-threaded, accelerated processor architectures in order to adapt source code assets at a similar pace?

new processor architectures

The processor ecosystem now evolves at a rapid and unpredictable rate in order to address the increasing demands of HPC users.

The fact is, source code assets simply cannot evolve at the same rate, especially when a new processor architecture is introduced every two to three years. Code modernization is required in order to succeed, however it’s a process that involves expert technical skills and tools. Data practitioners specialize in developing algorithms and applications, and typically do not have a keen understanding of the latest processor architectures or the desire to become experts in their own right. Quantitative analysts, or quants, want to benefit from accelerated compute power without entering the learning curve of the latest processor architecture.

The industry is responding with tools such as the HPE Quantitative Finance Library solution, which is designed to help developers modernize application software by generating highly parallelized source code for multi-core, multi-thread platforms. By outsourcing non-core code optimization activities, quants can devote more attention to improving algorithms and save the time and costs associated with learning the latest processor architecture.

New HPC tools are purpose-built for the needs of financial services and can allow coders to focus on innovations that drive revenue, instead of modifying existing code. Cost-effective and simply-to-deploy solutions can help financial firms achieve business growth, enhance employee productivity, and increase agility through HPC innovation.

This guest article was submitted by Suresh Aswani, Senior Manager, Solutions Marketing, at HPE. Follow Aswani on Twitter @sureshaswani2 for more tips on accelerating quantitative analysis and news on HPE’s latest HPC solutions for the financial services industry.