Grid Computing Adoption in the Financial Services Sector
The need for computing power within the financial services industry has long been a critical issue. Increased competition, a changing regulatory environment and economic pressures have led many financial institutions to find that their existing computing resources are no longer sufficient to meet demands. These demands include conducting complex tasks such as market analysis, pricing and risk management which are essential if financial institutions are to outpace competitors and make the best possible decisions in the shortest space of time. Access to computing power that is both scalable and reliable has quickly become an important requirement for the financial services industry. This has led financial services firms to deploy grid computing solutions.
Grid computing is technology that allows users to run complex, often process or data intensive applications on clusters of commodity servers. The technology offers organisations a great deal of flexibility and reliability by creating a system of shared resources that are able to compute large or complex tasks in a fraction of time that a single computer would take by distributing the calculations across hundreds or thousands of monitors. The concept of grid computing first emerged in the mid-1980s after scientific organisations and government labs began to require greater computational power. Since then, grid has been widely adopted for commercial use within a number of industries including financial services, which was quick to recognise its huge potential and benefits.
These benefits are certainly compelling. In an industry in which time to market is critical, grid computing plays an important role. For example, before the introduction of grid computing calculating potential outcomes for risk management scenarios was very time consuming since the entire simulation was run on a single, costly server. The process was very expensive as sourcing additional processing power was achieved by adding more hardware capacity to a single department. The introduction of grid computing has changed this by enabling tasks to be broken down and distributed across cheap commodity computing resources. As a result, increasing the scale of computational power is inexpensive and easy, and the time required to run trade analysis applications has been dramatically reduced. Since grid deployments allow companies to make use of their unused computing resources, it also helps to reduce physical IT resources and costs while maintaining a flexible and reliable infrastructure to cope with business demands.
While large-scale capital markets firms have been enjoying these benefits, the majority of the financial services industry – such as banking and insurance – remains behind the curve in the extent of their adopting the technology. Whereas companies in other sectors such as electronics and life sciences are utilising grid for a range of R&D activities, financial institutions lag behind. This is largely due to organisational culture.
Most financial companies typically separate organisational and IT infrastructures by department and tend to keep their IT information close to their chests. This approach to organisational structure results in grid being used for limited activity such as risk management applications and ignores its wider potential. If financial organisations are to release grid enterprise-wide there needs to be a change in IT practices, the development of new skills and a shift in internal policies.
If only the industry remained static, this might not pose such a serious threat to the laggards, however, as the leading companies in the financial services industry continue pushing the envelope, falling behind becomes dangerous. Today, some financial institutions are building the capability to perform intra-day pricing and risk analysis, which gives them a real-time understanding of the risk and valuation of the transactions they are making. Such intensive calculations can only be effectively performed with the computational power enabled through grids. As the deployments of grid solutions have matured, the argument for extending and sharing grid resources across the organisation has become a natural technology evolution, enabling more individual departments to cost-effectively meet business demands.
Some firms have already begun to take these steps and are adopting grid for an increasing number of tasks. This adoption process can take time and typically involves four successive stages. In the first stage a single application runs on a cluster of commodity servers or workstations. These commodity clusters can lead to a significant competitive advantage, since a need for more power is easy and inexpensive to accommodate by adding nodes to a rack in the cluster rather than the many-month process of adding an expensive SMP server.
In the second stage of grid adoption, organisations expand the size of their clusters and spread implementation to more departments and locations. Most importantly, this stage introduces a more efficient sharing of resources. Users begin running multiple applications in the same cluster which leads to increased monitor utilisation rates. Through the effective sharing of resources applications are able to get the power they need at the time they need it, while other parts of the organisation are able to access power that would otherwise go to waste. Some planning is required as the ideal scenario is to match up applications that experience peak usage at different times of the day.
This stage is often met with some resistance within financial organisations. Traders, for instance, are often hesitant to share their computing resources with others such as risk management. In order to tackle these internal issues, resource sharing policies need to be introduced. Examples of such policies include priority scheduling, fair share policies and time-based resource allocation for day/night and weekday/weekend.
The third stage of grid adoption sees resource sharing spanning the whole organisation as the operation of the grid shifts from individual lines of businesses or locations to central IT. This stage also explores the benefits of sharing expensive software licenses along with increasing computing power. This setup is often referred to as a shared resource or utility computing model. The organisation’s separate lines of business pay for usage, rather than ownership to access computing resources for their applications. This stage often requires a significant philosophical shift, particularly within financial institutions, as users are required to share resources that they have traditionally always owned and controlled. IT departments also adapt as they extend beyond just running the corporate network and adopt the responsibilities of running computing allocation.
In the fourth stage of grid adoption, enterprises seek to expand their infrastructure past modelling and analytical applications to make grid an integral part of the data centre. Here, standard business applications such as customer relationship management and human resources go on the grid. Thanks to a transparent view into internal usage patterns and the ability to highly customise service delivery, organisations can more easily prioritise resource allocation and make more informed decisions regarding when to access external services.
Grid computing offers organisations a range of possibilities. The financial services industry has not been as quick as other industries to realise the wider potentials of grid technology. This, however, looks set to change as organisations increasingly adopt an enterprise approach to managing their IT resources. This change in mindset is essential if companies are to enjoy the full range of business capabilities offered by grid and embark upon the four stage process of adoption. The need to improve business efficiency and cut costs has become more important than ever before. After all, seven minutes is better than seven hours. Seven milliseconds is even better. Grid makes this possible.