What most people do is put agents on machines monitoring the hardware and agents to collect log data. Those worked great in the days of monolithic applications, but in modern architectures that use distributed applications and possibly ephemeral containers, the data may not even exist — it’s virtual and ephemeral!
Moreover, applications have changed because infrastructures have changed. Applications are no longer just sitting in one location. They’re distributed. If an application is distributed with hundreds of components in distributed systems and possibly containers, finding out which components are running (as it may float among containers) in which container or VM, makes it very difficult to troubleshoot.
We started this piece about finding the trade. Good luck also finding the container without modernized approaches.
Finally, some of the compliance checks for trade settlement may involve querying third-party APIs, such as, “Is the customer on a sanctions list?” What if the third-party API is having latency issues or responding very slowly. The trade settlement process still has to continue. Continuously monitoring partners’ APIs for potential problems becomes just as important as monitoring your own software and hardware.
A modernized approach is critical — and incorporates observability
We realize many companies have already started the process of addressing T+1. In that same spirit, we would like to add to the discussion for making the effort to become T+1 compliant and more resilient. As compliance rules like T+1 evolve, we have to have a modernized approach — visibility and problem solving need to be almost as instantaneous as the trades themselves. Humans can’t handle it alone. Especially if the problem is a big outage because of something like bad code or denial of service. We might be able to figure out one outage. But what if there were many? The stakes are too high to leave it to chance, or rely on virtual war rooms. If we have half the time to solve a problem, we have to figure out the root cause and remediate using the most cutting-edge monitoring capabilities out there.
A modern approach incorporates observability. Observability is used to discover assets and provide instrumentation, identify active components, their performance and health, and contextual knowledge on the system. The single pane of glass that can monitor any trade in the settlement process and handle scale is the solution. A bonus is also observing the security of all trade settlement assets and their associated identities’ behavior, who may have access to the assets, adds to the resilience of the modern approach that is required for T+1.
It’s just too much for humans to solve at once. The faster we detect when an issue hits, the faster we are going to solve the problem.
