May 19, 2008
When Platform Computing announced that it formed a Financial Services Business Unit, it signified not only the company's evolving position on the industry, but also an evolution of the financial services industry itself.
From Platform's perspective, the justification of a dedicated business unit is simple enough. For starters, Platform has doubled its industry growth since increasing its financial services investment last spring, so, said Founder and CEO Songnian Zhou, it just makes sense to have a streamlined management team that can focus on banking-specific technologies and sales without going to the CEO or executive team to make every decision. Additionally, says Zhou, it can be difficult for a company to address various markets with the same solutions and personnel. An industry like financial services is especially difficult in this sense because the applications are different and the customer demands are far different, he said -- and they continue to demand more.
For one, says Zhou, although financial firms still do a lot of batch processing, there has been a "massive movement" over the past year to expand grid use into the real-time realm. Real-time processing for pre-trade pricing and analytics is much more mission-critical and demanding than overnight processing, says Zhou, so "customers become a lot more demanding in availability, reliability [and] in capacity in the surge." Customers are making or losing money in sub-seconds, he added, so they need power when they need it. He contrasts this usage model with that of the automotive industry, where the development cycle can span the course of a year or more.
Platform already has put a lot of effort into reducing latency times for real-time processing. The company's Symphony middleware was designed to handle low-latency tasks, said Zhou, with roundtrip transaction time in the 2-3 millisecond range. Times can be reduced even further by taking advantage of Symphony for IBM's Cell processor, which allows users to offload specific jobs or portions of jobs that can best take advantage of Cell's capabilities. ... "If you don't get to below 10 to 15 milliseconds, you can't meet the requirements of this kind of pre-trade [environment]," said Zhou.
Jim Mancuso, general manager of the newly formed Financial Services Business Unit, said both companies (Platform and IBM) are working with banks as potential beta customers to evaluate the Cell solution, and they expect it to have a "bright future." Although it's early, he added, Platform has a functional product and IBM is excited to show off the potential of applications on the Cell processor.
Aside from focusing on technologies to enable real-time analytics processing, Zhou said the Financial Services Business Unit also will focus on emerging trends designed to improve flexibility, reduce costs and reduce energy consumption for increasingly cash- and power-starved financial firms.
Among these trends is the seemingly omnipresent cloud computing. Zhou believes that banks initially will use cloud computing as a way to get extra power during peak demand times, but that it will become more pervasive over time. The reason for this interest is not only the promise of extra computing power, but also the promise of a reduction in electrical power.
However, while both of these attributes are of high importance to financial institutions, Zhou warns that cloud computing providers must make sure two issues are addressed before they can expect to see significant uptake from this industry. First, they must make sure that running applications in the cloud is not dramatically different than running them on an in-house grid. "Second, I think, is cost," explained Zhou. "A number of enterprise customers in financial services have done an excellent job driving down the cost per CPU-hour. ... The cost has been driven down from a dollar-plus to around 40 cents per CPU-hour, so the service providers need to make sure their offerings are more cost-effective than that."
Of course, Zhou doesn't envision a future where middleware vendors like Platform are irrelevant. He sees Platform serving the role of a "software arms dealer" in the cloud market, providing software to manage both internal and provider clouds. Serving both sides, he says, will help to make sure the process of utilizing cloud resources is transparent and seamless between internal and external datacenters. "We are providing software like we are selling groceries," analogized Zhou. "We can sell to the individual households who buy the stuff from us and cook at home, or we can sell to the restaurants. In some cases, people can buy our software and then [make an agreement] with the cloud provider to run that software on the cloud provider's site. ... We need to be open to all of these possibilities, because it's not one size fits all."
But cloud computing isn't the only trend the financial world is turning to save on energy; financial firms also are seriously looking at "green" IT. Zhou is quick to note, though, that when it comes to energy efficiency, everything is relative. For example, while running more applications on more nodes might not sound green, getting utilization rates above 90 percent epitomizes making the most of what you have. However, Zhou noted that utilization is just the beginning. Platform is working on technologies to dynamically power on and power off servers, and Zhou says advanced reporting and analysis tools allow even the most efficient datacenters to be ran smarter.
Given the difficulties the financial services market has experienced recently, it shouldn't be surprising that firms are working to everything "smarter." Zhou says one of the biggest lessons investment banks have learned is that they have to be more careful in how they run their businesses, especially in regard to pricing and risk analysis. “[T]hat requires a lot of computing,” he said, "but you cannot do it in a dumb way." With this in mind, he says the grid is more critical than ever as large financial institutions like JPMorgan and Citigroup attempt to increase flexibility and save money by establishing division- and company-wide services-based grids. Managing one larger, more inclusive grid is cheaper than managing 10 line-of-business clusters, and the shared services model allows for business applications to join computing applications on the high-performance infrastructure.
"It's ironic that while our big customers are going through a lot of these challenges in the market, they are aggressively expanding the deployment of grid technologies to deliver shared services because they have proven great ROI," said Zhou, "but also they are seeing ongoing, dramatic expansions of computing demand [necessary] to run their businesses responsibly."
In contrast to large customers getting more aggressive with their grid initiatives, second-tier banks and hedge funds around the globe also are upping their adoption, but they are demanding simplicity. According to Mancuso, a recent meeting of Platform's financial advisory council confirmed that smaller clients without large IT staffs are more concerned with just being able to get on the grid rather than with advanced usage models. To this end, says Zhou, Platform is working with systems partners like Dell and HP and OS vendors like Red Hat to provide integrated, holistic solutions (hardware, OS, middleware, scheduler, data cache, etc.) that will allow these smaller banks and hedge funds to run their applications and go on with their business without investing too much effort on IT management.
Jun 19, 2013 |
Ruan Pethiyagoda, Cameron Boehmer, John S. Dvorak, and Tim Sze, trained at San Francisco’s Hack Reactor, an institute designed for intense fast paced learning of programming, put together a program based on the N-Queens algorithm designed by the University of Cambridge’s Martin Richards, and modified it to run in parallel across multiple machines.
Jun 17, 2013 |
With that in mind, Datapipe hopes to establish themselves as a green-savvy HPC cloud provider with their recently announced Stratosphere platform. Datapipe markets Stratosphere as a green HPC cloud service and in doing so partnering with Verne Global and their Icelandic datacenter, which is known for its propensity in green computing.
Jun 12, 2013 |
Cloud computing is gaining ground in utilization by mid-sized institutions who are looking to expand their experimental high performance computing resources. As such, IBM released what they call Redbooks, in part to assist institutions’ movement of high performance computing applications to the cloud.
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/02/2012 | AMD | Developers today are just beginning to explore the potential of heterogeneous computing, but the potential for this new paradigm is huge. This brief article reviews how the technology might impact a range of application development areas, including client experiences and cloud-based data management. As platforms like OpenCL continue to evolve, the benefits of heterogeneous computing will become even more accessible. Use this quick article to jump-start your own thinking on heterogeneous computing.