October 29, 2015 | Written by: Andrea Braida
Share this post:
IBM BigInsights is an enterprise-ready Apache Hadoop® distribution designed to enable analytics on larger volumes of data more cost-effectively. In part 1 of this interview, we sat down with Rohan Vaidyanathan, the offering manager for BigInsights on Cloud, to discuss major benefits of the cloud service. In this installment, we’ll talk about how companies are actively using BigInsights on Cloud today to tackle their big data projects. In part 3, Rohan and Katherine Franklin, program director for security and compliance, share some important security developments around BigInsights and other Cloud Data Services offerings.
Hi Rohan! What kinds of businesses are using BigInsights on Cloud?
Over the last few months, we’ve had some very successful use cases with a variety of customers. We’re working with banks, insurance companies, pharmaceutical companies, and all of these companies have different needs and goals specific to their data.
How do their needs differ from company to company?
To just give you a sense, I’ll give you three use cases.
One is an automobile company that is streaming data into BigInsights from computers onboard cars. That data is then being analyzed to provide information on driver behavior, where the drivers are geographically, things like that.
BigInsights is being used in conjunction with [IBM InfoSphere] Streams and Cloudant, another IBM Cloud Data Service, to provide a solution which can build real-time analytics. So, imagine someone is driving around a city block looking for a parking spot. We can send them alerts pointing them to a free spot. That’s one of the use cases that we are driving towards (no pun intended).
This company is also using the technology to analyze user behavior in an attempt to find better insurance rates. That’s a combination of real-time analysis and model building, and perfecting that model on BigInsights, again in combination with Streams and Cloudant.
You mentioned two other use cases?
There is also a pharmaceutical company working on electronic medical records along with some additional data using R algorithms. Their users are primarily statisticians and mathematicians who know much of R, but they’re not familiar with Hadoop and the management of it. For example, they wouldn’t know much if anything about setting up a Hadoop cluster. The value that BigInsights brings to them is that we take care of all of the aspects of managing and running the Hadoop environment. They’re completely focused on computing data and using Big R to execute the machine learning algorithms using our frameworks on these electronic medical records to discover patterns within them. Big R makes it easy for users familiar with R to work with data hosted in BigInsights.
So that’s an out and out data science use.
And then the third customer we have is a large bank migrating one of their main data warehouses from on-premise to the cloud in Hadoop. In this data warehouse, they have a lot of mission critical systems, so we are making use of Hadoop to augment the existing data warehouse and gradually move it over to the cloud. In this way, we allow the bank to take advantage of cloud economics as well as the other IBM Cloud Data Services offerings hosted in the cloud. They can perform a wider variety of analytics in the same place.
Why is the bank migrating their data warehouse to the cloud?
A few reasons. A lot of data is born on the cloud. A lot of their customer-facing conversations are happening on, for example, CRM tools which are hosted on the cloud. They might be communicating with their customers, looking at customer feedback through social media assets. Moving all of the data from these sources to on-premise doesn’t really make sense. Data born on the cloud is best to stay there and then analyze.
Also, it’s useful to bring their customer data to the cloud so they have it all in one place. It enables a wider variety of analytics.
There’s the economics aspect – it’s cheaper on the cloud, cheaper to expand your footprint. You also have more flexibility to either scale up or down based on what you require in your business.
Finally, security, with which users are getting comfortable. Cloud based services are are squarely focused on security in an attempt to provide better business value. IBM is committed to security of our cloud services. We are focusing a lot on standards compliance, not just in terms of the software or the features we build into it, but also across the organization and how people interact with customers’ data. Our external communication channels are SSL encrypted and have options for dedicated VPN. There is encryption for data-at-rest. Clusters are within individual private networks. There are very strict controls over how and who interacts with the systems on which customer data is hosted. So that brings more confidence to our customers.
So a lot of exciting use cases. A lot of different use cases. From different industries. And all of these are already executed or in the process of being set up on BigInsights on Cloud.
Learn more about BigInsights on Cloud or get started on IBM Bluemix today.