Balancing Peformance and Simplicity for Wall Street

By By Derrick Harris, Managing Editor

October 29, 2007

Designed with Simplicity in Mind

Although the financial services industry is known for its early adoption of cutting-edge technologies – a task made easier by the high skill level of many of the industry’s IT personnel – even the most advanced users appreciate simplicity as they begin to expand their use of distributed computing beyond its traditional HPC realm and expose it to a more mainstream pool of developers and system architects.

In this regard, some would say it doesn’t get much better than GigaSpaces, whose version 6.0 technology, dubbed eXtreme Application Platform (XAP), has been designed with developer friendliness and ease of use as top priorities. Said GigaSpaces chief marketing officer Geva Perry, the company took a traditional programmer’s point of view when developing the new solution, which means no new APIs or programming models. Instead, thanks to its use of the Spring Framework, application writing on the GigaSpaces platform is transparent, with developers basically just writing Java, .NET, C++, or whatever code they prefer.

Perry hopes this will help open the doors to a wider audience whose members don’t necessarily have the unique skills of the Wall Street crowd. “They’re the smartest guys in the industry that you will find as developer and architects; they can … wrap their heads around [high technology],” he said. “I don’t think that in financial services we need to do much more as vendors for increased adoption.”

Of course, as noted earlier, it isn’t just the distributed computing masterminds who are working with these technologies nowadays. According to Marc Jacobs, a director at financial services application developer Lab49, most of the non-specialists to whom he has spoken are just coming to terms with distributed computing techniques, so the easier the better. They just want to “take [an] app that’s already written in serial mode, tweak it, do a couple of things to it and get it distributed,” and GigaSpaces makes this a reality.

Another element of simplicity is holism, and this is another area where GigaSpaces excels. XAP innately handles data, business logic and messaging, which leads to one of, if not the, most complete solutions on the market. Perry is quick to point out, however, that GigaSpaces’ application platform is not just an amalgamation of parts sewn together, but rather a product of a different approach. The company’s founders and chief technologists thought about the problem “like computer scientists” trying to design a solution from scratch (as opposed simply trying to improve upon existing solutions) and the resulting core technology yields a single product instead of one big product that is really three components.

Jacobs appreciates the holistic qualities of the GigaSpaces platform, as well — especially its distributed caching and automation capabilities. Competitive solutions handling business logic, for example, suffer from not having a distributed cache. As for automation, Jacobs likes the how abstraction XAP creates allows developers to write objects and dump them into a distributed cloud, then simply call the methods on the objects and have them distributed across multiple machines. Developers also can combine the data cache along with a mechanism for scheduling method calls on different machines.

Nevertheless, and although he believes the company is leading the pack in terms of including a collection of technologies and features that are “destined to all come together,”  Jacobs acknowledges that the GigaSpaces platform does have its setbacks. In fact, if asked everyone’s favorite clichéd interview question, he might well suggest GigaSpaces respond with everyone’s favorite clichéd response: “My biggest fault? Probably that I try too hard.” In Jacobs’ opinion, GigaSpaces is guilty of “going for reach rather than polish” by continuing to add new features instead of refining its existing pieces. He also thinks the company might have bound its hands by sticking with the JavaSpaces API on which it is built. That standard, he noted, is not really designed to handle everything GigaSpaces is trying to do, and the end result is a product that feels “a bit kludged.”

In addition, Jacobs said that the “people who write the checks” care more about the management side of things than developer simplicity (“They care more about security than they care about a good API.”), and GigaSpaces is relatively weak on that end when compared to solutions from vendors like Platform Computing that contain very rich management and security capabilities.

For his part, GigaSpaces’ Perry doesn’t delineate so clearly the line between his company and the Platforms of the world. From GigaSpaces’ perspective, its product works very well as a complement to the traditional, industrial-strength grid platforms – they handle resource coordination and big batch jobs, and XAP handles the real-time, data-intensive applications. GigaSpaces already has partnered with Platform and DataSynapse, and is working closely with Microsoft and Sun Microsystems to increase interoperability with Windows Compute Cluster Server and Sun Grid Engine.

Overall, however, “They are … visionaries,” says Jacobs. “They actually recognize much more so than their competitors that the distributed computing platforms of the future are going to have lots of different pieces in them … and they need to be envisioning that sort of holistic platform, not as a bunch of best-of-breed products that I have to figure out how to weld together.”

Simplicity Draws Commerzbank to XAP …

Among GigaSpaces’ stable of financial services customers is Germany’s Commerzbank, who earlier this year began developing its new Risk Data Framework on top of GigaSpaces XAP.

Designed to improve the credit risk evaluation process and make it possible for the bank to do more and better business, Risk Data Framework will calculate the risk of counterparty portfolios with a Monte Carlo simulation using Sungard’s Adaptiv Analytics product. Before the simulation process is triggered, however, all of the trade data has to be parsed, transformed, aggregated and mapped to a proprietary API call. In order to have all of these tasks carried out in a timely manner, Commerzbank chose GigaSpaces.

According to project manager and Java architect Sebastian Titakis, who is responsible for the conception and implementation of solutions for the business unit in charge of credit risk monitoring, GigaSpaces’ simplicity was a big help in his quest to improve time-to-market of the new application via increased developer productivity. “Thanks to the simplicity of the GigaSpaces API and its integration with the Spring Framework,” he said, “we were able to develop very fast.”

Titakis added that the support of languages like Java, .NET and C++ fits into the IT landscape of many financial organizations.

The inherent functionality of GigaSpaces XAP’s aforementioned holism also proved a big hit with Titakis and his team. One key he noted along this front was being able to concentrate on developing a domain model with its business logic without having to write in any plumbing in the infrastructure layer. The business logic was not polluted and they were able to increase code reusability and testability.

In fact, Titakis did not even look at any other grid solutions when evaluating his options, but he did look into some other architectures, including a batch application with a database backend, and message-oriented architecture. In the end, though, none of the alternatives could satisfy all of the requirements like GigaSpaces “space-based architecture,” nor could they match GigaSpaces in terms of development time and budget-friendliness.

Basically, GigaSpaces offers a very flexible, as well as a simple, grid model,” explained Titakis. “If you compare the grid functionality of GigaSpaces with other solutions or architectures, you will have to invest considerably more time to reach the same level of functionality, especially in terms of disposing the required expertise and dominating the technology. Learning to use GigaSpaces is easy and really does not take much time for an average programmer.”

But Don’t Forget about Performance

Simplicity aside, however, Commerzbank also chose GigaSpaces to free its credit risk analysis operations from their legacy database-centric architecture. For one, XAP delivered immediate performance gains thanks to its in-memory data caching, as the business processes have access to the necessary data and business logic without having to make a call to the database. In addition, the product’s distributed nature and failover capabilities mean Commerzbank doesn’t have to worry about losing valuable trade data.

Looking to the future, Titakis said Commerzbank is expecting trade data volumes to grow by 10 to 25 percent annually, and the ability to linearly scale the bank’s GigaSpaces architecture just by adding additional blades will have a big impact on its ability to handle that data. He added that the solution’s inherent low-latency also will enable near-future evolution of the Risk Data Framework, which originally was designed to run every evening but soon will become an event-driven application that triggers real-time Monte Carlo simulations. Additional projects will be developed on top of the GigaSpaces architecture whenever the business case justifies it.

There are many aspects that make GigaSpaces XAP 6.0 a very interesting product for data grid solutions,” Titakis commented. “I think it is the right product if you need to develop event-driven applications that require low latency and high transactional throughput. Its ability to scale, as well as its high availability, failover and transactional support, makes it perfect for financial institutions.

Real-Time is the Future

To hear GigaSpaces’ Perry tell it, Commerzbank is not alone its quest to achieve maximum scalability, availability and overall performance, all while limiting latency to the lowest possible levels. GigaSpaces receives about 50 percent of its overall revenue from the financial services industry, and Perry says most, if not all, firms are concerned with “extremely low latency.” And, perhaps most importantly for GigaSpaces and its distributed-architecture-peddling cohorts, the financial world is all but settled on the fact that traditional database and J2EE application server approaches won’t cut it anymore.

Among the trends ratcheting up the urgency for low latency are new regulations, such as Regulation NMS, and the emergence of algorithmic trading, where machines are buying and selling from other machines. In regard to the latter, “Latency has become such a critical issue that even if you don’t necessarily have a better price, you might get the business if you’re able to trade a millisecond faster than the guy across the street,” says Perry.

The end result of these trends, as well as an overall increase in business, has been exponentially growing data and transaction volumes, which, in turn, have led to an increased reliance on commodity hardware. Whereas in the past organizations could buy a big high-end server and, assuming data and transactions were grew linearly, rely on it to suffice for a few years, today’s financial services firms have to deal with volumes that are, in some instances, quadrupling every year. In the face of this kind of pressure, Perry says organizations have turned to distributed computing platforms running on cheap servers, where scale can be increased very quickly and with zero to minimal code changes.

However, simply switching to a distributed architecture doesn’t necessarily solve all of an organization’s problems. In today’s highly transactional, real-time financial datacenters, said Perry, applications need constant and very fast access to data – a problem which GigaSpaces has addressed with it distributed, in-memory data caching capability.

On top of this, an application platform has to make sure the right data is located with the appropriate business logic, a process some call “affinity,” and something Perry said GigaSpaces has figured out. “[T]here may be five or six or 12 events that are related to that same transaction,” he hypothesized, “so I have to make sure the data related to that transaction is always in the same place as the events that are coming in that are related to that transactions.”

As noted earlier, though, GigaSpaces is looking to expand its presence outside of Wall Street, an area in which the company has grown quite comfortable since its inception. One area where Perry sees real potential is Web 2.0, as online retailers and service providers are facing similar pressures in terms of transaction volumes and latency requirements, and the leaders in this space already have proven the need for distributed platforms. “They’re looking more and more like the Wall Street firms,” said Perry. For GigaSpaces, that’s a good thing.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

Nvidia Appoints Andy Grant as EMEA Director of Supercomputing, Higher Education, and AI

March 22, 2024

Nvidia recently appointed Andy Grant as Director, Supercomputing, Higher Education, and AI for Europe, the Middle East, and Africa (EMEA). With over 25 years of high-performance computing (HPC) experience, Grant brings a Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Houston We Have a Solution: Addressing the HPC and Tech Talent Gap

March 15, 2024

Generations of Houstonian teachers, counselors, and parents have either worked in the aerospace industry or know people who do - the prospect of entering the fi Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire