To address market changes rapidly and enhance productivity, enterprises need to harness data in new ways. Chris Swan, vice president and chief technology officer for the Global Delivery organization at DXC Technology and a DXC Fellow, talks about applying analytics, lean processes and automation for efficient and effective operations.
Q: Why do so many organizations undertaking a digital transformation need additional help?
A: One thing that comes up time and again is network connectivity — and its security. Also, organizations generally have to find some way of connecting their cloud environment to their existing environment. They’ve got several choices, but each comes with trade-offs on cost, performance and security. This leads to the shared-responsibility model, where chief information officers (CIOs) essentially get a service provider to take some of their risk.
Q: So service providers accept some of the risk while also helping the organization get more productive, right?
A: Yes. For example, we’re capturing, processing and storing much larger datasets than we did in the past, but at a cost that’s much smaller than before. Now we’re mostly processing the data in flows that exist outside of traditional relational database management systems (RDBMS).
We’re also storing data in object storage, which is orders of magnitude cheaper than putting things into an RDBMS. We then use those object stores for machine learning training. We can also use them to ask questions we forgot to ask in the first place. They’re optimized for a completely different approach.
An RDBMS is very good at answering queries, whereas object storage is optimized for the cost of storage. It’s really cheap to store a lot of data, and there’s a different emphasis. So we don’t expect to be running lots of queries on our object storage because we’ve already extracted the data value on its way into the object storage in our data processing flows.
Q: Can automation help, too?
A: Yes, and this connects with what we’ve been doing with DXC Bionix™, our methodology for improving the efficiency and effectiveness of operations, which is done through analytics, lean and automation. Platform DXC™, which underpins Bionix, is used to optimize delivery of our managed services and is also integrated into our next-generation offerings.
When customers buy those offerings from us, they’re getting Platform DXC and Bionix built into what they have. We use the phrase “powered by Bionix.” That means this analytics-based approach and the optimizations we make from it are baked in from the beginning, rather than coming along later.
DXC itself is a consumer of these services, so we know what that’s like, and it’s driven us through our own cultural change. We’ve had to shift our own ways of working, moving away from waterfall to agile, from a project-centric approach to a product-centric approach, from traditional Information Technology Infrastructure Library (ITIL) mechanisms for delivery to ITIL-enabled by DevOps pipelines.
Q: What benefits can organizations expect from an automated, data-driven approach?
A: We focus on four key metrics:
• Lead time: How long it takes to get from a customer request to something being done about it. We’re looking to deliver systems that reduce lead time massively.
• Deployment frequency: How often we can get a change into the environment. Some CIOs are happy going from two or four times a year to weekly. But in many cases we’re actually trying to do deployments daily or even hourly.
• Mean time to restore: How long it takes to recover from a failure. You have to anticipate that failures are going to happen. So you want to minimize the time it takes to bring the system back up when those failures do happen.
• Change-fail percentage: When we make a change, how often does it not actually work out, so that we need to roll it back and stick with the environment we had? Our goal is to drive toward every change being successful. So that’s a proxy measure for the quality of what we’re putting through our process.
Overall, the business is asking for continuous transformation. DevOps gives them continuous learning by experimentation. Every time we deliver something new into the environment, we gain an opportunity to learn whether it fits. Then the product can evolve to its environment. Ultimately, that’s how we support continuous transformation: with IT systems — from infrastructure all the way up to applications — that continuously adapt to the environment and its changes.
To learn more, read the white paper, Use IT modernization to accelerate and scale business transformation.