Organizations must take incremental approach to incorporating data with core

Register now

While there’s a lot of attention on digital platform delivering the next generation of insurance services, many insurers still maintain what may be called legacy infrastructures, with mainframes and relational database management systems underpinning operations and transactions.

Will they eventually move away from these systems? The answer may be: only if necessary. It’s likely the future will be built on a combination of existing underlying assets, with new-age technologies built on top.

That’s the prognosis from a new survey of 1,200 IT and business professionals, conducted and released by Progress. While Progress has a horse in this race – it offers mainframe modernization solutions – the survey is nonetheless telling in that it suggests many enterprises remain committed to building on their existing legacy infrastructures. It can be assumed that these results also reflect the mainframe-intensive insurance industry. At the same time, the massive influx of data poses unprecedented new challenges to organizations, which may not be quite ready to handle it all.

The survey finds, for example, that relational databases are still going strong. Sixty-three percent pull in data from Microsoft SQL Server, 45% from MySQL, and 44% from Oracle.

At the same time, cloud continues to become a major piece of the enterprise IT landscape. Software as a Service adoption increased from 62% to 79% since last year, the survey’s authors report. More than 50% of respondents use two or more SaaS data sources, while more than 35% use three or more. This rapid growth in SaaS adoption is primarily driven by CRM, HRM and marketing applications. Of these, the top data sources are Office 365, Salesforce and Google Analytics, which increased the most (11%) since 2016.

To meet the looming data influx, enterprises are turning to cloud and open source platforms to manage it all. Hadoop Hive is the most popular big data interface and is expected to grow 8% in the next two years, the survey finds. In addition, new, cloud and big-data-friendly database types, the NoSQL class of technology, gains steam, with more than 50% using NoSQL, with the pen-source MongoDB being the most popular.

The number one integration challenge is an increasing number of data sources, the survey also finds. The leading issue is data spread across an increasing number of data sources (cited by 49%), integrating cloud data with on-premises data (41%), and managing data Variety involving both structured and unstructured formats (32%).

Rising SaaS and NoSQL adoption, combined with continuing significant investments in on-premises relational database systems, “creates a major challenge in accessing on-premises data from the cloud with increasing complexity in firewall policies mixed with expanding data protection laws and security standards,” the survey’s authors caution. “Using the cloud requires that they seamlessly integrate the two environments. However, many voiced concerns about accessing that data from the cloud. Traditional technologies for connecting these database silos are difficult to manage and neither scalable nor engineered for the cloud.”

The survey also pointed to standards being adopted for data connectivity. From 2016 to 2017, use of ODBC increased 11%, JDBC 11%, OData 3%, while REST increased significantly by 19%. “Each standard is closely linked to different trends in data consumption across various shapes, sizes and locations. Big data sources have been driving adoption towards SQL standards such as ODBC and JDBC. And the OData REST API is gaining ground in hybrid and cloud environments because of its rich set of interoperable query capabilities to unify data access for disparate resources across the Internet.”

The survey’s authors talk of the need for greater “hybrid connectivity,” especially with the rowing convergence of cloud and on-premises data. “Growing SaaS adoption and the necessity of on-premises data (both relational and big data) highlight the need for real-time connectivity across the entire hybrid architecture.”

However, they add, “connectivity between SaaS platforms and on-premises systems behind the firewall is a complex issue. And security vulnerabilities, data privacy and compliance are top challenges. To overcome the challenge of exploding enterprise data sources and increasing SaaS adoption, the on-premises and cloud systems must be connected in a seamless way. Hybrid connectivity leverages open standards to easily connect the expanding hybrid environment to get the most value out of the data.”

The bottom line: as cloud grows, so will the work that needs to be done to make it all click.

For reprint and licensing requests for this article, click here.
Cloud computing Data warehouses