Make way, make way…Dodd-Frank is coming through. Signed into law 2010 it is now becoming a reality for Wall Street. Not an IT issue, you say? Guess again. This highly impactful law will require ever more vigilance from IT to ensure that their businesses are able to meet its stringent requirements for reporting and transparency.
Some of the highlights of this regulation include the following: all swaps need to be reported as soon as “technologically practical”, confirmation messages need to be reported with thirty minutes of execution, rejected trade messages must be submitted within the original thirty minute deadline, all trades must be retrievable for up to five years after maturity or termination and much more. These requirements are not easily accomplished and have a big impact on IT. You definitely don’t want to be reactive and miss one of these deadlines and “hope” to do better next time. Being compliant will require the utmost in proactive application performance monitoring to ensure you make these short windows.
So, what do you do? Start rationalizing all the overlapping tools you have and ensure what remains all contributes to a single “version of truth” for your datacenter. If you haven’t already done it, ensure that you have automatically maintained end-to-end views of your key applications impacted by these requirements. Remove blind spots where you have insufficient monitoring. And most importantly, be able to trace your trade transactions in near real-time and so you can take action as soon as there is an issue and before users are impacted. Often the root cause of problems will be sudden upswings in market volume and the inability of your infrastructure to handle these without queue backlogs. Avoiding these is key to ensuring reporting compliance.
To make this work, you will need to be able to recognize a problem at the business transaction or process level such as a reconciliation failure or a DTCC reporting failure so you can take appropriate action. This requires analytics such as complex event processing to rapidly identify activity that is impactful to your processes and not merely a low level alert.