MapR recently announced the release of MapR-DB 6.0, as part of their Converged Data Platform. This release coincided with Convergence London, a one-day conference focused on showcasing the real-world value of enterprise data, which is part of the global Convergence series of conferences led by MapR. As a close MapR partner with a tight, certified technical integration into MapR’s architecture, Arcadia Data was invited to participate in the conference to share our expertise on big data. As a regional event, it felt inclusive and members of Arcadia Data met with numerous technologists, demonstrating that while there are a lot of organisations out there either providing software or services, or indeed industry expertise in the big data space, there is simply no one-stop-shop. There are so many ways to leverage big data that no single technology can provide all the answers to all the questions being posed of organisations looking to exploit all of their data. When it comes to big data, one needs to consider many different components that address specific needs. One great example is how you handle historical data versus real-time data. One might have once assumed that you put real-time data into the same store as your other data and run queries there. However, businesses today are recognizing the introduction of streams-based technologies to complement their existing data stores such as RDBMs, data warehouses, and Apache Hadoop.
Convergence London featured a variety of MapR customers and partners as presenters. As an example, Nick Hough-Robbins, the chief data officer at Specsavers (a global optical retail business), spoke about the future of the company and its move from being a commodity glasses and lens manufacturer and retailer, to a medical provider. This business transformation is very much supported by data if not led by it. Specsavers currently has data from 100+ million glasses prescriptions, and is now able to identify future ailments such as Alzheimer’s, dementia, and Parkinson’s using Ophthalmic Coherence Tomography scans more accurately than human experts, which in my opinion is a great use of data. So rather than just sticking with the dataset they had, Specsavers embraced the spirit of big data and explored more data sources to create new opportunities.
The one technology-focused MapR presentation of the day, presented by Anoop Dawar (VP of Product Management), was particularly interesting as it focused on the move toward data-driven applications. Today, most organisations that have adopted the latest data technology platforms are focused on the technical aspects such as getting data into the systems, ensuring data quality and provenance, and getting data and insight to the end users. The move toward data-driven applications, based on all that data goodness that supports existing and new business processes, is the next step. In the near future, simply providing dashboards and KPIs won’t be enough. Applications that are designed and created by the business users for the business user will be not only required but expected. This self-service application creation is an area that Arcadia Data sees as the next step in data visualisation.
In my talk, I shared my perspectives on big data, and a topic that always seems to generate a lot of interest is how to derive value from big data for business users. Sure, there are a lot of great tools for technical folks, but what about the rest of the analytics-focused community who are looking to leverage new, large data sources? With all the industry talk about “data scientists are rare and expensive,” businesses need to plan ahead on how they’re going to make big data usable for the large, existing groups of end users. Part of that planning entails technology, and part of it is process, but fortunately Arcadia Data has a technology that simplifies some of the hard parts of the process.
One key area is with regard to optimizing analytic responses. In other technologies, this is done with OLAP cubes, and is sometimes done today with extremely expensive, dedicated analytical engines. I discussed the Arcadia Data approach, in which our flagship product, Arcadia Enterprise, provides a feature called “analytical views” that allows massive scale and fast responses with much lower IT effort. These analytical views are pre-computed aggregations that speed up a wide range of queries. So they act like OLAP cubes but avoid the long cycles of determining which cubes one needs to build. Arcadia Enterprise even has a built-in recommendation engine that suggests the types of analytical views to build, so you can get optimizations with much less IT intervention. This is only part of our overall story around data agility, so I encourage you to learn more from our website, especially in the Resources section.
As an event, I think Convergence London highlighted that in EMEA, organisations are slowly moving away from thinking about on-premises Hadoop as a data lake for batch processing and moving toward a hybrid distributed computing architecture. Such an architecture combines on-premises and cloud technologies and in most cases include components of the Hadoop ecosystem. This seems to mirror the evolution occurring in the United States specifically with reference to projects like Spark and streaming data capabilities that are becoming more prevalent.
All in all, it was great to meet some new potential clients and partners, and to see external validation for the path that Arcadia Data is taking. Please reach out to us as you pursue your data-driven initiatives, as we’d love to hear what you are looking to do, and we’d be happy to share our insights that can get you going.