For some time, IBM’s Cognos analytics solution has been a popular choice for enterprise analysts and data teams looking to uncover crucial business insights from their company’s growing datasets.

While Cognos has performed its role well, many data engineers and analysts are waking up to the fact that their Cognos solution isn’t keeping up with the datasets and queries their business is working with in today’s big data environment.

So, what happened?

OLAP Technology and the Cognos PowerCube

If you’ve never heard of OLAP (OnLine Analytical Processing), you probably don’t remember the siloed and limited data analytics environment that preceded it. At the time, it was a big deal. OLAP broke down these limitations and helped data teams consolidate their datasets and quickly perform more scalable and complex analytics. Since then, business data has become ubiquitous, and OLAP analytics continues to help business analysts generate more sophisticated insights.

For IBM Cognos users, PowerPlay made this OLAP approach possible for organization’s everywhere with its introduction of PowerCubes. Combined with Cognos’ powerful visualization tools, IBM’s OLAP PowerCubes provided a powerful business intelligence environment for all but the most data-intensive analytics work.

OLAP and Big Data’s Growing Demands

Unfortunately, that data-intensive analytics work is now commonplace in organizations of all sizes across every industry. Insights from Big Data are now seen as a major strategic advantage, and businesses are sparing no expense hiring an army of data scientists, analysts, and engineers to collect all the data they can and mine insights faster than ever before.

This has put organizations that run on Cognos in a bind. Cognos is built into their analytics workflows and their analysts prefer its visualization and reporting features, but it’s just too much work to manage and too slow when they run it on their massive datasets. What’s going on here?

As it turns out, a few things. First, PowerCube size and dimension limitations have begun to matter. This wasn’t the case years ago. Now, with queries becoming more detailed and datasets scaling into the petabytes, PowerCube size constraints are a liability.

The response to this has been for organizations to build hundreds (or thousands) of PowerCubes, but this isn’t without its own drawbacks. Now, data engineering teams are tied up building new cubes and analysts are forced to wait for cubes to render.

This render time means analytics insights are always a day or two behind real-time, and Cognos’ limited self-service analytics beyond 10 dimensions means analysts either have to settle for surface-level insights or spend more time waiting for answers to their detailed queries.

Speaking of queries, query speeds can barely keep up in the terabyte range. With datasets often reaching into the petabytes these days, query speed limitations are all too common. And if you’re looking at historical data (e.g. Year on Year), latency becomes even worse. In short, organizations that adopted Cognos PowerCubes to stay ahead of the competition in the early days of Big Data are now finding themselves falling behind right as Big Data is taking off.

Kyligence OLAP Engine – A Path to the Future

The good news for Cognos users is that there’s an easy way to protect much of their Cognos analytics investment while solving the underlying problems that plague it today. Kyligence’saugmented OLAP platform (based on Apache Kylin’s extreme OLAP technology) enables businesses running on Cognos to replace and consolidate their PowerCubes into a single massively multidimensional cube. This provides several major benefits:

• Faster Insights and Query Speeds – Kyligence enables sub-second query speeds on massively multidimensional cubes that render in a couple of hours instead of a couple days. This means analysts can get to work and get results with as little downtime as possible.
• Self-Service and Greater Interactivity – With limitless cube dimensionality and no coding required for data prep, data modeling, and data analysis, analysts can ask more detailed questions and dig deeper into the data with no major dependencies on engineering.
• Reduced Development and IT Costs – Consolidate all of your PowerCubes into one easy to manage and render Kyligence OLAP cube.
• More Flexibility for Enterprise Analytics – High scalability and concurrency, rich API support, and granular data access controls.

All of this comes together to provide Cognos analytics users a faster path to new insights and less management overhead for their data engineering and IT teams.

For example, China’s UnionPay saw their TCO drop by nearly 90% when they transitioned their Cognos-based data warehouse over to one built on Kyligence. This was thanks to the considerable decrease in hardware and software costs, as well as the reduced manpower from eliminating the maintenance of thousands of OLAP cubes and data preparation scripts.

Get Started with Kyligence OLAP for Cognos Analytics

If you’re done playing catch up with your analytics insights and want to learn more about how Kyligence can bring your Cognos deployment into the 21st century, download our Cognos Solution Overview.

Curious about OLAP? We recommend checking out the recent presentation from Kyligence CEO, Luke Han:

Augmented OLAP for Big Data from Luke Han

And if you want even more information, this video offers a great overview of the Kyligence and Apache Kylin approach to augmented OLAP analytics:

Also, be sure to follow us on LinkedIn and Twitter for the latest Kyligence product updates andAugmented Analytics announcements.