More organizations are prioritizing big data analytics initiatives and transforming their operations. Yet building a modern data analytics platform is not a straightforward task. Many organizations struggle to extract insights, or evolving analytics to become a business discipline.[1]
A data analytics platform should offer both an architecture and a product that allows many different business users to extract value out of data. It must be able to evolve with the business to support and drive value creation that can align to future strategies.
Building our big data analytics platform has allowed Western Digital to embark on a great journey to leverage data to drive business decisions across the organizations. In a video interview, JuneAn Lanigan, Western Digital global leader of Enterprise Data Management shares some of the insights we’ve gathered over the last six years and why futureproofing is a key concept you should adopt as step one in approaching your data analytics platform design.
This blog is part of a series sharing insights on Western Digital’s big data analytics platform.
Video Transcript:
Hi, my name is JuneAn Lanigan and I’m the global leader of Enterprise Data Management at Western Digital. I’m very excited to be here with you today, sharing the story of our journey over the last six years to build out a fully scalable big data [analytics] platform and ecosystem.
We’ve learned a lot, taken a lot of risks, and definitely pushed ourselves into territories we were very unfamiliar with in order to accomplish what we now have is a best-in-class Big Data architecture and ecosystem. That ecosystem supports and provides value to our product development, our engineering, our manufacturing operations, as well as our product quality teams. We’re very excited about that, but let me first set the stage with where we were when we began.
Data Analytics Platform – Where We Began
Our hard disk drives are built across 16 different factories, so there are a lot of silos of data, silos of knowledge, and there’s silos of perception around what data strategy is, what analytics is, and the type of analytics that people were doing. The goal really was to bring all of that together into a common platform that was accessible to all stakeholders, but was also able to provide value to those stakeholders at the stage of their lifecycle — where they fit in the [lifecycle of the] build of a drive.
The other thing that’s interesting is that we conduct 6,000 tests on every single hard disk drive. All of that data needs to be connected. Our vision was really to create the DNA of the drive – to be able to put in a serial number and see everything about the drive, including the customer field experience data so we could truly understand the drive [from manufacturing through its lifetime].
The Challenge
The challenge was ‘how do we do that’ (create the DNA of a drive)? It’s one thing to just bring all the data into a data lake, but the other is to how to build up those views that absolutely support that integrated view, as well as make sure that we have five nines of data quality, data latency, etc.
Developing a “Futureproof” Strategy for Data Analytics
So, let’s move into futureproofing because that’s really what the goal was for us. What we mean by “futureproofing” is [that] our data strategy not only can meet the needs of the business today, but as the business matures and evolves, the analytics capabilities that we’re able to provide in the platform can support that as well.
We’ve been on this journey together with the business. Everybody has been hands on to help figure out how to use this data analytics platform to the benefit and bottom-line advantage of Western Digital.
Where we look at futureproofing is kind of in three main major areas:
- It’s All About the Data
The first one is around the data itself. We are really thinking about what kind of data we need to be able to accommodate on the platform. You can think about test data, sensor data off machines, or you can think about images. It’s really about pushing the envelope of the type of data that we will need and aligning to the future strategies in the business around industry 4.0 or smart factory. Those are common terms in the industry for us.
As we look at all of this, it’s not only the type of data [we need], but also how do we process that data? [For example] you process image data very differently than you process sensor data, and so you have to think about what tools and technologies and techniques do we use to incorporate that data into our data analytics platform.
- Creating Value
Next [it] is really about value – how do we derive business value from this data? We have to cater to the type of user that’s a stakeholder, as well. You have some people who are more apt to just want to run a dashboard, or run a report. Or maybe they want to do ad hoc, they want to drill into the data. But, what we’re finding more and more as we mature, what we call the digital DNA of Western Digital, now, is that we’re looking at machine learning and artificial intelligence. We’re trying to move from just being reactive to truly being predictive and ultimately to the point where we want to be prescriptive. That’s what I mean by futureproofing.
- Flexible IT
Looking Back to Look Ahead
[To summarize], when I talk about futureproofing you have to think about a lot of things. You have to think about where you’ve been, you have to think about the culture of the company today, how its evolving and what the needs are going to be. You [also] have to think about the different technology capabilities that exist today and those that are emerging, and how those fit. You also have to think about the type of access people need to the data, and the tools they need to do the type of analysis that actually brings value to the bottom line.This blog is part of a series sharing insights on Western Digital’s big data analytics platform. Subscribe to the blog to be notified when new blogs are published.
Learn More
JuneAn also shares her insights in a webcast interview: Lessons Learned from a Data Management Expert
[1] https://www.techrepublic.com/article/big-data-adoption-exploding-but-companies-struggle-to-extract-meaningful-information/