Free Word Search


Search by Topic

  • Keyword
    Industry
    Purpose
    Expert
    Area

HOME NRI JOURNAL The Insurance Systems in the Cloud Computing Age

NRI JOURNAL

Innovation magazine that generates hints for the future

クラウドの潮流――進化するクラウド・サービスと変化する企業の意識

The Insurance Systems in the Cloud Computing Age

Akira Matsumoto, Senior Managing Director and Deputy Division Manager of the Insurance Solution Division

#Management

May 07, 2018

Up to now, Japanese insurance companies have built most of their core systems on mainframes, and these systems have long supported the businesses of insurance companies through their robustness and stability. These companies have amassed IT assets over a long history, developing legacy systems in the process. Many insurance companies thus have a strong awareness of the challenges involved when it comes to the personnel resources and costs required for systems maintenance and management.

Insurance Companies Facing a Rush to Migrate Their IT Assets to the Cloud

As one means of solving these challenges, some have suggested going “post-mainframe”—that is, streamlining your IT assets into an open-source or cloud-based platform, and then transferring them to a new system. However, companies that try streamlining their IT assets will face an additional challenge, namely the difficulty of significantly rebuilding a large and complicated core system.

Incidentally, it’s worth asking whether mainframes can continue to support the businesses of insurance companies into the future. According to a survey conducted by the Japan Electronics and Information Technology Industries Association (JEITA), shipping records for mainframes have steadily declined since fiscal year 2000 in terms both of volume and of monetary value, and are now a tenth of monetary value in 2000. Of course, this doesn’t mean that mainframes will be going the way of the dodo in the immediate future.

Nonetheless, with the mainframe market now shrinking, manufacturers are being weeded out of the competition and the market is almost certainly on the way to becoming an oligopoly. If that happens, it’s possible that costs in general will remain high while levels of service will typically decline. Meanwhile, the cloud computing market is growing by leaps and bounds, and the full-scale use of cloud platforms has already begun in the banking and securities industries as well as in other financial sectors.

Going Post-Mainframe Requires Nearly Ten Years to Complete

Given these circumstances, what options do Japanese insurance companies have? For answers, let’s turn our attention to the U.S., which is known for being an insurance and IT superpower.

In August 2017, NRI conducted face-to-face interviews with representatives of several leading U.S. insurance companies on the topic: “Thoughts on Going Post-Mainframe and Legacy Systems in the Age of Cloud Computing”. All of these companies have mainframes, and Japanese insurance companies could probably stand to learn much from them. While each company’s circumstances were different, they were all in agreement on the necessity of actively leveraging the cloud, and given the costs and migration risks, they all believed that going post-mainframe would require a nearly ten-year period to complete.

Further, the motive of these U.S. insurance companies for turning to cloud computing wasn’t simply a matter of cost reduction. Even more than that, they feared the depletion of their mainframe-related personnel.

NRI Researching Insurance Systems for the Cloud Computing Age

NRI has contemplated what form insurance systems should take in the digital computing age from two perspectives: IT strategies and migration methods.

When it comes to IT strategies, NRI believes that given what the future holds for mainframes, companies need to create ten-year action plans for their core systems and implement projects as soon as possible. The first year should be spent setting your objectives for the next ten years—that is, establishing what your in-house systems should be, including for instance your IT platform, development system, and IT personnel strategy in light of your overall business strategy. Then in the following year, you’ll want to take inventory of your company’s current assets, and get a good grasp of how your systems are performing, as well as think about your migration method and assigning levels of priority. Use the subsequent period for actually building your systems and executing the migration. But remember, you’re envisioning the entire project to take nearly ten years to complete, so it would be best to pursue this endeavor under the strong leadership of your top management.

Now, let’s look at the two methods available for migration: “Lift-and-Shift”, and “Wrap-and-Renew”.

Lift-and-Shift is the method most strongly advocated by U.S. cloud vendors. This method entails “lifting” your existing system configuration and applications with minimal modifications onto a cloud platform, and then gradually “shifting” your system constituent elements and applications to a cloud service. It’s best suited to systems in non-core IT domains—in other words, systems that can potentially leverage the cloud in their current state sooner rather than later.

The second method, Wrap-and-Renew, involves “wrapping” together and using both your new and old systems for a certain period, in order to reduce your risks. With this method, even after adopting your new system, you would store your old one with the aim of eliminating the data, and then once you’ve reduced it to a certain scale you would “renew” it by migrating it over to the new system. This method is probably best suited to systems in core IT domains such as contract management, that is, high-risk systems that in their current condition would be difficult to reconstitute or to migrate wholesale.

NRI is investigating what insurance systems need to be in the cloud computing era, focusing on the post-mainframe phenomenon, the different uses of cloud vendors depending on the purpose, and the migration methods for achieving such purpose.

 

NRI Research Paper Knowledge Creation and Integration March, 2018

  • Facebook
  • Twitter
  • LinkedIn

What's New