Author: Suvo Dutta

Cloud trends that will define 2020

What’s your strategy for these “cloud” trends in 2020? 1. Multi-cloud- Not one cloud platform but the “best of the breed” approach will prevail. Customer-facing apps, business processes, analytics, & APIs running in disparate cloud platforms & seamlessly connected would be the new “normal”. 2. On-premise apps will be “truly” modernized to cloud-native apps for the real benefits of cloud adoption. AS-IS Lift & Shift, or data-center migrations alone are not likely to work anymore. 3. Re-skilling would remain high, not just for technical teams but also for business, PMs, sponsors, etc. since “cloud” transforms not just technology but also...

Which business function is the biggest factor for customer retention for insurers?

Which business function is the biggest factor for customer retention for insurers? Well, what can it be if not billing! As policyholders how frequently do we look into marketing materials or even the policy document? How often do we submit claims, or call up the insurer? But, can we escape from looking into the premium bills, paying them, or asking questions about them! We can’t & so billing guarantees customer touchpoints that no other functions do. That makes it the insurer’s trusted gateway to its customers. A great billing experience can make a difference. Over 70% of insurers agree that...

Should we convert an existing Data Warehouse to Data Lake?

Should we convert an existing Data Warehouse to Data Lake? Often I get this question. I think this decision should be driven by rational evaluation of business needs & not because “all cool dudes are doing it”. DW has been around for a couple of decades. Historically, it’s been a gold-copy of operational data in structured format & the main BI enabler. Many of these DWs are built on legacy platforms on-premise with layers of interconnected data-marts evolving over years into complex “bus” constructs. Also, the business that used a DW for decades understands its data very well for analytic...

Which 3 ways AI is likely to disrupt the Insurance industry at a massive pace?

Which 3 ways AI is likely to disrupt the Insurance industry at a massive pace? Recently I attended a workshop, that addressed how AI is disrupting the insurance industry, with the keynote that while the insurance industry was always slow in embracing transformation, that’s not true anymore. I was also reading through this article that coincidentally addressed a similar point. Today, venture capitalists understand the huge potential for technologies like AI to revolutionalize insurance. There are likely 3 potential areas where AI can bring “change” to an industry that is known for its stiffness & phobia to change – 1....

A serverless approach to relational data replication!

A serverless approach to relational data replication! I felt it challenging keeping cost & complexity at bay while designing real-time replication of data from operational RDBMSs, primarily due to the need for layering in multiple middlewares like CDCs, message brokers, message streams, etc. in different capacities which often result in a lot of plumbing. Also, the commercially available middleware platforms are not always free of compatibility issues, integration complexities & high costs. An alternate serverless option that I found handy is to leverage MySQL-compatible Amazon Aurora for such replications by leveraging its Stored Procedure that invokes Lambda functions. While it...

Is IT the driver or order-taker?

Is IT the driver or order-taker? I came across this interesting article, the link of which is at the end of this article, that looks back through the IT evolution over the last decade about how that once perceived to be true & infallible became overtly irrelevant today! What were the best practices a decade ago are roadblocks now! Here are some major ones. A decade ago it was IT that ran the show & business did as IT instructed, but today it is Business that runs the show & IT must deliver what business asks for. Back then IT...

Nine best practices for microservices deployments

What best practices do you follow in deploying microservices in AWS? Assume an AI/ML model needs to be deployed in AWS & exposed as an API endpoint. What are the best practices? Is any guardrail available? While not a standard procedure & not limited to these, here is what I follow at minimum when I deploy an app service, for example, a Python ML model in AWS. Please share your thoughts. 1. Ensuring the right VPC for network isolation 2. Containerized hosting of the app service in ECS cluster spun in private subnets for access control 3. ECS cluster spanning...

Which 3 things make digital transformation the biggest concern for insurers, even more than the sluggish economy & low-interest rates?

Paper assets – Over decades, insurers accumulated paper assets that could stretch for miles. Most of those are confidential & must be archived for regulatory reasons. They require special storage & authorized access. Thus, managing these paper assets is cost-intensive & time-consuming which makes operations inefficient. At the same time, the digitization of paper assets of that scale is nearly impossible. Regulations – Insurers run in a highly regulated environment. There are stringent guardrails in trying new ideas, & restrictions in the use of PII, PCI, & PHI data. These limit capabilities in a lab environment which slows down innovations....

What 4 things if not addressed can kill your microservices strategy?

Microservices provide freedom from the clutches of the “spaghetti monster” and lets us to the world of flexibility, scalability & agility, but they don’t come free. Here are 4 reasons why! 1. While microservices reduce the “impact” of IT changes through the separation of concerns they can increase the “frequency” of changes that can impact business in the absence of a right level of DevOps maturity, the severity of which rises exponentially as they grow in number. 2. Microservices are founded on the principles of “separation”. This means “isolation” is important in their management from people, process & technology standpoints....

What workloads you must move to the Cloud – Part 3 (for Batch workloads)

Batch workloads date back to the early days of computation when mainframes dominated the computing world.  They still play a significant role in different disciplines including business, engineering, medical, healthcare, and other areas. Batch workloads are normally designed to run in the background and are meant to process huge amounts of data. What are the challenges of running batch workloads on-premise? 1. Poor customer experience Less availability of business functionality On-premise IT ecosystems typically have batch workloads that need an extensive amount of computing and storage provisioning. This is why real-time workloads cannot share environments with batch workloads due to...