Data then, data now

5 ways cloud data warehousing as a service is changing the future of data warehousing

Product Manager, IBM

Think back to the types of analytics you created for your organization three to four years ago. Now fast forward to today: spot the differences?

Most corporations have seen a massive change in the types of analytics created for business—and the amount and types of data collected and governed. The world of data is changing: we are well beyond the time when data was only limited to core financial and business operations reports.

Today, data transcends departments; combined information now drives new, actionable analytics for sales, service centers, marketing, human resources, business operations and beyond.

We’re facing a data-driven future where a real request might sound something like “Analyze orders to find growth trends, and then overlay social and web data to learn the sentiment about our product so we can use these results to anticipate demand in our manufacturing process.”

Business users want this information quickly, and they want it to be hyper-specific and accurate. How does an IT professional deliver?

The answer might be in flexible deployment options.

Cloud-based data warehouse solutions can complement and extend traditional data warehouses. Pairing the traditional data warehouse with a cloud-based solution adds speed and flexibility, helping you “do more with less” and add business value.

The Data Warehousing Institute (TDWI) recently created a checklist report for cloud data warehousing which call out some of the benefits of a cloud-based approach:

  • the deployment platform to the analytics purpose: look at the type of analytics and choose the right deployment platform. When needs are ad-hoc, limited time or seasonal, self-service or for prototyping, a cloud data warehouse is a good option. 
  • Shorten time to value by simplifying deployment, where cloud providers do the infrastructure and provisioning work for you—and they may even provide advanced analytics capabilities for you. Let them handle all of these tasks for you, so you can focus on providing value through requirements and innovation, where your goal is to be as specific as possible about the analytics themselves. 
  • Look at the analytics, where it can be a time-saver to leverage those that are integrated into the solution. Be sure they suit your analytics project. For example, the dashDB fully managed service includes in-database analytics that have been proven by enterprise customers, in addition to, open source R analytics and Spark analytics as well. 
  • Look for consistent performance of the solution, where performance must be good enough to encourage adoption, success and eliminate the headaches that go along with a dissatisfied user base. We all know the importance of performance, so understand if there are flexible options and what is included in the underlying platform software and hardware to ensure good performance. In-memory computing and in-database analytics are aspects of this.
  • Watch data security and governance needs, where you must look at access control, encryption for data at rest and in motion, and also the ability to mask sensitive data. 

While there are many other considerations, these are some of the key points that stood out to me from this report. Take a look at it and please share your conclusions and thoughts.