Software Testing Seen as Crucial to Restoring Confidence12.24.2012
Trading technology, including algorithms, is undergoing a facelift as the industry seeks to plug holes that resulted in the Knight Trading snafu and other fiascos this year.
It’s borrowing a page from large software developers, whose systems must undergo rigorous testing before an undiagnosed bug can cause damage.
“Just like any well engineered system, every link in the chain for ‘the system’ needs to support the intended goal and check for anomalies at every step,” said Scott Ignall, chief technology officer at Lightspeed Financial. “The system in this case is the trading firms, technology providers, broker dealers, and exchanges. They all need to provide the necessary pre-trade checks to prevent these errors.”
In the aftermath of the May 6, 2010 “flash crash,” the SEC and the exchanges worked to devise new protections to keep computer trading errors from spreading too rapidly or inflicting unacceptable harm on the overall market.
The exchanges reformed their rules for breaking trades, instituted single-stock circuit breakers, updated market-wide circuit breakers, and implemented limit up/limit down mechanisms.
One important area of focus is testing and industry preparedness.
“The industry has learned through experience that we must change the way we test,” said Eric Noll, executive vice president and head, Nasdaq OMX Transaction Services, at a hearing on computerized trading held last week by the Senate Banking Committee.
In the past, industry-wide system changes have utilized a testing methodology that tested for system design integrity.
For example, “we might test a software update by having our members send us test orders to ensure that the software does what we are asking it to do,” Noll said. “Instead, we should be testing each other’s systems to try to break them.”
A more robust testing environment would assume breakdowns by all testing participants to visualize the impact on a system’s integrity.
“Such ‘destructive’ testing will spot troubles that the kinder-gentler testing of the past would not uncover,” said Noll.
While regulators have done a good job of tackling individual components, the entire system needs to be viewed holistically when analyzing the risk of these systems.
The Volcker Rule, Large Trader and the Market Access Rule (SEC 15c-3) are having profound impacts on trading technology.
“All of these rules present significant technology challenges to front and back office systems,” said Ignall.
Large Trader in particular “seeks to pierce the veil behind trading technologies to discover who individual traders are,” Ignall said. “The problem is the interlinkages between trading systems, brokers, exchanges often don’t carry this identifying information. Many of these linkages were actually built with the explicit goal of obfuscating the individual trader for fear of front-running and information leakage.”
Many industry experts are warning the untangling these hard-coded protections will present a massive effort costing the industry an overwhelming amount of money – both real dollars, and opportunity cost.
“We would prefer the market regulators focus on making the CAT (Consolidated Audit Trail) as robust as possible so it can also handle the large trader surveillance in conjunction with CAT’s other stated goals,” Ignall said. “This would allow the industry to focus on one ‘catch all project’, rather than several smaller, yet very intrusive projects.”
Nasdaq is partnering with Carnegie Mellon University to help bring the industry together to improve the resilience of financial services technology.
“We hope to form a group of market participants, regulators, technology providers and academic institutions with the goal of driving resilience in the large-scale software engineering and technology arena,” Noll said.
Strategy Studio allows brokers to create their own execution platforms.
Allowing algorithms to control the supply and demand of crypto was challenged by Terra Luna.
Infrastructure and sustainability will be one of the biggest opportunities in alternative investments.
The broker's algo development team gains access to high quality, granular Level 3 data.
Bank aims to gain share from its systematic approach to trading, market structure proficiency and modern techn...