Meet Your AI Copilot fot Data Learn More

Why OLAP Persists in the Era of Big Data Analytics

Yasmin Criss
Technical Writer
Aug. 30, 2021

Data is the lifeblood of businesses and institutions everywhere. It allows for more informed decision-making and, ultimately, improves organizational processes. However, with so many technological advancements, an organization might find itself swimming in an overabundance of data. 


Figures collated by TechJury show that people produced around 2.5 quintillion bytes of data every day in 2020, with each person creating 1.7 MB of data per second. Given this, organizations are challenged to extrapolate only the most valuable insights from huge amounts of data. Often, this is done with data and analytics tools — one of the earliest being online analytical processing (OLAP). To this day, OLAP programs are utilized for business intelligence. This article will delve into why that's the case.

A Brief History of OLAP


OLAP is used to arrange large databases and put the data through rigorous analyses. Though a robust application now, earlier versions used to take the form of rudimentary spreadsheets with special data-related features.


The very first OLAP product was much like this. It was known as the Express and was created in the 1970s. Following its release, the industry saw the creation of even more powerful spreadsheet-like applications. They had features for database connection, multidimensional modeling, statistics recording, and more. These features would later become staples in the modern OLAP.


Later, large technological corporations, such as Microsoft and Oracle, created their own OLAP software and released it to the industry. Since then, OLAP solutions have been widely in use across the industry. Though more advanced data and analytics tools have begun cropping up, seeming to take their place.


Now, data professionals are proclaiming that OLAP is dead. But this conclusion is largely misguided. As discussed in our article "News of the Death of OLAP Has Been Greatly Exaggerated," OLAP is based on a sound theory that has proven beneficial throughout the years, across different industries. That said, haphazardly concluding that OLAP is at the end of its rope is premature at best.



One of the biggest reasons why OLAP persists despite its supposedly "outdated" nature, is that it is largely more accessible to laypeople. To illustrate, consider data lakes. Whereas OLAP produces structured data based on user queries, data lakes store structured, semi-structured, and unstructured and uncategorized data. This allows for more analytical potential, but it also makes for a more complicated analytics process. As such, this alternative is more suited for experts who have advanced knowledge in data engineering and data management.


Because data lakes have been growing in popularity over the past years, so has the demand for learned data professionals to navigate them. This has not only led to a spike in jobs but also the courses available to train data specialists. In particular, higher education institutions have opened up their data programs at every level to online students, and are just as comprehensive as traditional degrees. 


Maryville University's online master's in data analytics focuses on looking for trends, making decisions, identifying opportunities, combining operational data with analytical tools, and presenting complex information - all key data skills needed in the workforce. And because of this opening up of online education, the job outlook for program analysts and operations research analysts is strong, with a growth rate of 11% and 25% respectively. And with the worldwide big data analytics revenue forecasted to reach around $274 billion in 2022, this demand will only continue to increase.


But the heightened demand for data professionals is an effect of a very specific phenomenon — namely, the growing complexity of data and analytics tools. While this is necessary given how the number of data sources and analysis methods continues to increase, it also makes data management more inaccessible to employees outside the data analytics field.


This is where OLAP shines. OLAP solutions are easy to navigate, even for non-experts. The interface is interactive, allowing for more data exploration, compared to static reports. And because data processing methods are already built into the model, employees only need basic digital skills to navigate the software. Thus, OLAP is not an outdated solution, but one among many. Though other advanced tools exist, data accessibility is still an integral consideration to organizations across industries.


OLAP and Big Data

Further proof that OLAP is not outdated is its continued adaption to accommodate modern data trends. Case in point: big data.


The benefits and uses of data science and big data are abundant. When combined, organizations can leverage huge datasets to produce more informed insights. This is a boon to the business industry as it extends the boundaries of data management and business intelligence. Given this, the importance of big data in the field is unquestionable. Moreover, OLAP solutions are very much capable of facilitating these big data processes. 


Researchers from the University of Alicante conducted tests on the open-source engine, Apache Kylin, and managed to create guidelines in optimal cube design for Big Data OLAP systems. They also laid down benchmarking goals to serve as key performance measures. These factors include building time, building success, and cube size, to name a few. The fact that OLAP software can be utilized to gather these big data insights is proof of its relevancy today. And it will continue to persist because, again, it is rooted in sound theory.





Case Study: Migrating from Cognos to Kyligence

Overview: Supercharge Your Cognos Analytics on Big Data

Blog: When 1,200 OLAP Cubes Become Two