For many firms, August 2007 provided an unpleasant view of the future. During what is traditionally a quiet trading month, volatility surged as the markets responded to an emerging credit crisis. Trading activity on the New York Stock Exchange soared 58% versus the previous seven-month average. At the same time, average trade size in August declined by 19% versus the previous seven months, creating a blizzard of orders and related market data that pushed firms' technology solutions to the limit. While things gradually returned to normal, many firms were left to contemplate a future in which such volumes were the rule, not the exception.

Legacy order management and trading systems under fire

August's trade volumes helped focus the market's attention on a serious problem facing firms today: legacy trade workflow and order execution technology. In a market environment where speed is measured in milliseconds and volumes have increased substantially, firms have to address the shortcomings of inflexible systems that simply cannot handle today's volumes and related messaging traffic. Many firms that rely on such platforms watched them slow to a crawl and, in some cases, collapse under the weight of August's trading activity.

In many cases, the primary culprit was the firm's legacy Order Management System (OMS). The reason? OMSs, introduced during the earliest phases of electronic trading, were never designed as execution platforms. Their stated purpose was to bring accounting, compliance, reporting and other similar functions under one umbrella. And given that only established asset management firms needed (or could afford) such a solution, OMS vendors focused squarely on the largest buy- and sell-side institutions. With this market in mind, OMS vendors adopted an 'everything but the kitchen sink model,' rolling out products that were heavy, complex and expensive, both from an integration and customization standpoint. As OMSs became deeply entrenched in firms' technology infrastructure, attached to a web of back and mid office applications (many of which were proprietary), client firms found themselves beholden to their OMS vendors for technology upgrades, consulting services and product enhancements. The cost of ripping out and replacing an OMS was prohibitive for most firms, and as such their OMS gradually became part of their legacy infrastructure.

Given OMSs' stated business model as internal workflow applications, combined with the market environment that prevailed during their initial rollout, it is not surprising that early OMSs lacked any trading functionality whatsoever. Unfortunately, even as the market for electronic trading evolved in response to structural and regulatory changes, OMS vendors' attitudes towards automated trading technology remained essentially unchanged. It was not until the early to mid 2000s when brokers began aggressively pushing their algorithmic strategies that OMSs focused on rolling out order execution functionality. By that point, legacy technology and rigid system architectures left OMS vendors unable to deploy anything approaching advanced trading functionality. Traders using OMSs for electronic execution could send orders to brokers' electronic destinations, receive confirmations, cancel and re-send, and, in certain cases, employ the most basic DMA functions. That, unfortunately, was the extent of OMSs trading capabilities, a model that persists to this day.

However, it was not just firms that rely on OMSs for electronic trade execution who suffered last summer. Many firms use a combination of OMSs for workflow purposes and broker Execution Management Systems (EMS) for trading. Compared to OMSs, broker EMSs represent a giant leap forward in trading functionality. However, like OMSs, many broker systems simply cannot handle the increased order flow and market data that accompanies periods of high volatility. Constrained by legacy architecture, these platforms were not designed to handle the kind of high frequency trading that many of their clients now demand. And because these systems are closed (hosted and managed by the broker in question), there is little that client firms can do to improve the situation.

In response, many firms who previously relied on legacy software installations are now turning to advanced, broker-neutral EMSs to address the inevitable increase in trade volumes. These platforms were built from the ground up to handle massive trade volumes and related messaging traffic. In many cases, they form the backbone of firms' in-house algorithmic trading systems, and as such were able to withstand the onslaught of market activity in August with no degradation in performance.