Microsoft



| | | |

| | |Windows Azure |

| | |Customer Solution Case Study |

| | | |

| | | |Financial Risk-Analysis Firm Enhances Capabilities with Dynamic Computing Capacity |

| |[pic] | | |

| | | | |

| | | | |

|Overview | | |“We’re using Windows Azure to meet the evolving needs of our clients. And with the … reduced levels |

|Country or Region: United States | | |of new investment required, we can consider solutions that might otherwise not have been possible.” |

|Industry: Financial analysis | | |Rob Fraser, Head of Cloud Computing, RiskMetrics |

| | | | |

|Customer Profile | | | |

|RiskMetrics provides risk management analysis| | | |

|for banks, investors, and asset managers. | | | |

|Headquartered in New York, RiskMetrics has | | | |

|more than 1,100 employees operating in 13 | | | |

|countries. | | | |

| | | | |

|Business Situation | | | |

|To enhance scalability of its services, | | | |

|RiskMetrics needed to support large bursts in| | | |

|computing activity over short periods of | | | |

|time. It wanted to empower innovative | | | |

|solutions by reducing the required new | | | |

|investments in its physical computing | | | |

|infrastructure. | | | |

| | | | |

|Solution | | | |

|RiskMetrics used the Windows Azure™ platform | | | |

|to provide on-demand computing capacity for | | | |

|its analytics applications. | | | |

| | | | |

|Benefits | | | |

|Elastic scaling | | | |

|Enhanced services, empowered innovation | | | |

|High reliability | | | |

|Improved agility | | | |

| | | |RiskMetrics Group’s Risk Management Business delivers risk management services to the world’s leading|

| | | |asset managers, banks, and institutions to help them measure and model complex financial instruments.|

| | | |To meet increasing market demand for risk analysis, RiskMetrics needed to accommodate increasing peak|

| | | |loads on its computing infrastructure for specific periods of time. The company used the Windows |

| | | |Azure™ platform—an Internet-scale services platform hosted in Microsoft® data centers—to provide |

| | | |on-demand computing capacity for its analytics applications. Now RiskMetrics can complement its |

| | | |on-premises capabilities with a flexible, reliable solution that can support bursts in computing |

| | | |activity over short periods of time, deliver enhanced services for more customers, empower |

| | | |innovation, and provide the company with increased business agility. |

| | | | |

| | | |[pic] |

| | | | |

Situation

RiskMetrics provides risk management, governance services, and financial research and analysis for the world’s leading investment banks, asset managers, hedge funds, pension funds, insurers, and central banks. Headquartered in New York, RiskMetrics has more than 1,100 employees operating in 13 countries. The company models over 4 million global securities daily, spanning 750,000 market data time series.

RiskMetrics delivers innovative technology solutions to help its customers better analyze risk and return. To estimate the risks associated with specific assets, financial engineers develop complex Monte Carlo simulation models, a class of algorithms used to generate and evaluate a broad range of estimated future values for each asset by projecting the risk factors that could influence the price of the asset over a range of time horizons.

Monte Carlo simulations require high computing capacity, and to provide a complete view of the risk factors influencing hundreds of customers’ financial portfolios, RiskMetrics may have to repeat the extremely complex simulations thousands of times for every customer—and do it within a restricted time window of a few hours. With a processing infrastructure that includes thousands of physical CPUs, the company provides technology services through a software-as-a-service subscription business model. It hosts its risk management applications in two large data centers in the United States and delivers them as services to customers across the world through the Internet.

As an early pioneer in financial software-as-a-service, RiskMetrics has the necessary computing power, management efficiency, and scalable multitenant architecture to deliver highly reliable risk management products and services. But to retain its leadership position in an increasingly complex market, the company maintains a heavy focus on research and development of new products and services to benefit its customers.

Given currently projected economic conditions and anticipated demands from enhanced financial regulation, market participants will require increasingly comprehensive risk analyses delivered on demand. To process that level of service, RiskMetrics will likely have to impose very high peak loads on its computing infrastructure for short periods of time, often demanding 10 times the computing resources required during other periods.

To provide even deeper, faster, more responsive, and more interactive risk analyses for its customers, RiskMetrics needed a way to support increasingly large bursts in computing activity. It wanted a computing solution that would enhance its business agility with reduced levels of new investment. At the same time, the company needed a solution that could meet critical standards for availability and reliability.

Solution

RiskMetrics decided to develop a solution that would host the processing of complex pricing calculations on the Internet through an external data center, a system often referred to as “cloud computing.” The company participated in a Microsoft® Technology Adoption Program for the Windows Azure™ platform, an Internet-scale cloud services platform that is hosted in Microsoft data centers. The Windows Azure platform provides an operating system and a set of developer services that can be used individually or together. Also, Windows Azure affordably fits the needs of customers by offering a scalable infrastructure with a pay-as-you-go model.

RiskMetrics already used the Microsoft .NET Framework 3.5.1 and the Microsoft Visual C++® 2008 development system to develop its existing applications. The company used those technologies to easily integrate its applications with the Windows Azure platform.

“The technologies we were already using are very compatible with Windows Azure,” says Richard Bower, Developer at RiskMetrics. “Working with the .NET Framework and with tools provided by Microsoft for the Visual Studio® development system, we can develop applications locally and deploy them to Windows Azure very quickly.”

Using Windows Azure, RiskMetrics developed a solution rapidly over the course of several months. To complement its on-premises infrastructure, RiskMetrics uses the Windows Azure cloud services operating system—the development, service hosting, and service management environment for the Windows Azure platform—to provide on-demand computing capacity for its analytics applications. During periods of high demand for specific kinds of complex analysis, the company can extend the capacity of its risk-analysis applications by deploying large numbers of Windows Azure resources to process the calculations. When the demand declines, RiskMetrics can scale down from the Windows Azure instances until they need the capacity again.

“We can take calculations we would normally do locally, process them on Windows Azure, and integrate the results with our local analytics architecture,” says Rob Fraser, Head of Cloud Computing at RiskMetrics. “We call the solution RiskBurst: bursts of processing onto the cloud from our risk infrastructure.”

RiskMetrics runs Worker role instances of Windows Azure to take messages from Queue Services and from the Blob Storage feature in Windows Azure. The worker roles pass on the messages to an external 32-bit calculation engine, which communicates with Windows Azure using Windows Communication Foundation. The company uses the Service Bus, part of Windows Azure, to monitor the messaging and the performance of the worker roles, measuring parameters such as CPU utilization, the number of messages in the queues and Blob Storage containers, and the number of transactions processed by the calculation engine.

With the Service Bus, RiskMetrics monitors the demand for processing power and adjusts the number of Windows Azure resources it uses to provide the processing capacity it needs. System administrators can provision additional Windows Azure instances in approximately 30 minutes. The company will fully integrate the solution into its existing operational infrastructure to allow monitoring of the messaging queues and dynamically provision Windows Azure instances as needed. Eventually, RiskMetrics expects to provision up to 30,000 Windows Azure instances per day.

The RiskBurst solution already relies on the communication between the RiskMetrics high-performance computing (HPC) architecture and Windows Azure, but the company anticipates developing increasingly seamless and scalable applications that span Windows Azure and Windows® HPC Server 2008, which is built on the Windows Server® 2008 operating system, to deliver both on-premises and cloud computing capacity as needed. “Potentially, we can use the right level of integration between Windows Azure and Windows HPC Server to provide seamless elasticity between local software and cloud-based services,” says Fraser. “It’s a very exciting combination of technologies.”

Benefits

With Windows Azure, RiskMetrics has extended its risk-analysis solutions to dynamically scale to meet processing demand. The company can now provide enhanced services for more customers, while reducing the amount of new investment required. It can innovate with more agility, bring new products to market that would not have otherwise been commercially viable.

Elastic Scaling

Using the vast computing and storage capacity in Windows Azure, RiskMetrics can enhance the scalability of its risk management applications, deploying large numbers of processing instances only when necessary. It can respond to demand as it comes, provisioning more instances of Windows Azure as required and reducing the number of instances as demand declines.

“That is not something we could have done in the past with just a fixed processing capacity,” says Philip Jacob, Head of Quantitative Development at RiskMetrics.

Enhanced Service, Empowered Innovation

With Windows Azure, RiskMetrics can now enhance the analyses of its customers’ complex financial assets. Using the vast computing resources of Windows Azure, the company can scale its solutions however it needs, empowering innovative new solutions.

“We’re using Windows Azure to meet the evolving needs of our clients,” says Fraser. “And with the significantly reduced levels of new investment required, we can consider solutions that might otherwise not have been possible.”

High Reliability

RiskMetrics has to meet binding service level agreements (SLA) to provide its services, and many of its customers cannot do business without receiving RiskMetrics analyses every day. The availability and reliability commitments provided with the Windows Azure platform are very important to RiskMetrics.

“We require binding SLA agreements from all providers we engage in our service delivery,” says Fraser. “By participating in the Technology Adoption Program for Windows Azure we gained insight and built trust in the people and engineering behind the SLA.”

Improved Agility

While deploying large scale dynamic computing capacity provides elasticity for the company’s RiskBurst analytics, RiskMetrics is also enhancing its business agility and empowering innovative services for its customers. By turning elements of its peak load infrastructure from a fixed cost to a variable cost, RiskMetrics can consider bringing new services and products to market that otherwise would not have been viable.

With the pay-per-use payment structure of Windows Azure, the company only pays for what it needs as it needs it, without having to make significant capital investments in fixed capacity, enabling RiskMetrics to be more reactive to market demands.

Windows Azure Platform

The Windows Azure platform provides an excellent foundation for expanding online product and service offerings. The main components include:

• Windows Azure. Windows Azure is the development, service hosting, and service management environment for the Windows Azure platform. Windows Azure provides developers with on-demand compute and storage to host, scale, and manage Web applications on the Internet through Microsoft data centers. In addition, Windows Azure serves developers’ connectivity needs through the following services.

← The Service Bus connects services and applications across network boundaries to help developers build distributed applications.

← The Access Control Service provides federated, claims-based access control for REST Web services.

• Microsoft SQL Azure™. Microsoft SQL Azure offers the first cloud-based relational and self-managed database service built on Microsoft SQL Server 2008 technologies. 

To learn more about the Windows Azure platform, visit:



[pic]

-----------------------

|“Working with the .NET Framework and with |

|tools provided by Microsoft for the Visual |

|Studio development system, we can develop |

|applications locally and deploy them to |

|Windows Azure very quickly.” |

|Richard Bower, Developer, RiskMetrics |

|“We can take calculations we would normally do|

|locally, process them on Windows Azure, and |

|integrate the results with our local analytics|

|architecture. We call the solution RiskBurst: |

|bursts of processing onto the cloud from our |

|risk infrastructure.” |

|Rob Fraser, Head of Cloud Computing, |

|RiskMetrics |

|“We require binding SLA agreements from all |

|providers we engage in our service delivery. |

|By participating in the Technology Adoption |

|Program for Windows Azure we gained insight |

|and built trust in the people and engineering |

|behind the SLA.” |

|Rob Fraser, Head of Cloud Computing, |

|RiskMetrics |

| |

− ww.

− For more information about RiskMetrics products and services, visit the Web site at:



− Additional Resources:

− Download: Windows Azure Tools and SDK

− View: Architecting and Developing for Windows Azure

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download