Managing the lifecycle of IT infrastructure components has likely never been more important. Deciding when to retire or extend the life of hardware such as servers, mainframes, network equipment and other large systems can have a huge impact on costs, energy consumption and how effectively IT delivers services to insurance firms and their customers.
"There are many factors that influence a carrier's decision to replace, refresh or upgrade hardware," says Martina Conlon, principal in the insurance practice at Novarica, a New York-based consulting firm serving the insurance and financial services industries.
Clearly, carriers want to stay current with their hardware to mitigate the technology risk of unsupported platforms, Conlon says. Upgrades and refreshes are also often driven by the requirements of applications that run on the hardware.
"By far, the most important consideration [in a hardware refresh] is the applications required to support tomorrow's business," says Tom Pettibone, founder and managing partner of Transition Partners, a Reston, Va., IT management consulting firm, and former CIO at New York Life Insurance Co. "Frequently those applications drive middleware and hardware decisions," Pettibone says.
In a few cases, the applications that are running on the hardware can indeed limit replacement options, Conlon notes. "Upgrading or replacing a major application can require a significant investment," she says. "If the carrier is not in a position to upgrade or replace the application, they can become locked into supporting obsolete hardware to keep the business running. Carriers have been known to shop e-Bay for older hardware or replacement parts so that they can extend the life of the platform to delay the cost and disruption of an application replacement."
Cost and Criticality
Another key consideration is the cost of maintaining and supporting hardware. "All these decisions take TCO [total cost of ownership] into account," Pettibone says. "At some point the cost of maintenance and the unreliability of older equipment exceeds the cost to upgrade."
At Penn National Insurance in Harrisburg, Pa., decisions on what to do with hardware often come down to budgets and priorities for how to allocate IT resources.
"As we plan for IT hardware replacements for the upcoming year, we may want to replace more equipment than our IT budget will allow," says Cindy Todoroff, director of infrastructure and client technologies at Penn. "If this is the case, we will make decisions about what equipment should remain in place and what equipment needs to retire. We may have equipment in place that is working well for us and is not causing us any issues even though it is aging. Depending on the criticality of the equipment we may decide to extend its life and not replace it."
Todoroff says in deciding whether to retire or extend equipment life, Penn considers a number of factors related to total costs and performance. One is whether equipment is still vendor supported.
"If the equipment has been 'end-of-lifed' by the vendor we will have concerns that we will not be able to get it repaired or get the replacement parts we need if we continue to maintain the equipment," she says.
Another factor is how reliable the equipment has been. "Has the equipment experienced failures as it has aged?" Todoroff asks. "If so we would not want to consider extending its life further."
Penn National tallies repair costs as equipment ages to determine whether it's cost effective to maintain the systems. "We may find that is less expensive to implement new equipment than to maintain old equipment," Todoroff says.
The firms also looks at its ability to repair equipment if problems arise. "We would not want to keep equipment in place if we did not feel we could not repair the hardware if it were to break," Todoroff says.
Conlon notes that the evaluation of a platform refresh and the total cost of hardware should involve an assessment of the platform's stability. For example, how many incidents have occurred in the past six months and whether the rate of incidents is increasing.
Other factors affect costs, Conlon says. "Renegotiating support and service contracts can often lead to significant cost savings, so significant that the initial capital costs can often be regained within a year or two," she says.
Many vendor options exist for commodity items such as servers and disks, so prices vary, tend to go down over time and are negotiable, Conlon says. "Carriers can additionally save on energy costs when migrating to newer platforms," she says. "Technology consolidation and virtualization can also drive hardware replacement decisions because they can reduce software licensing costs, simplify the technology architecture and reduce the infrastructure maintenance effort."
To help with the decision-making process on hardware, insurers can deploy IT asset management, which can include business practices as well as software applications that help manage technology assets. IT asset management can help companies make more informed decisions about hardware, gain greater control of burgeoning IT infrastructures, reduce risk and save money.
Pettibone says the technology choices are greater than when he managed New York Life's hardware infrastructure in the 90s, when the software was emerging but the process was still primitive.
"Basically we were still 'tagging' hardware and manually maintaining lists of software," Pettibone says. Since then, management products have been enhanced to enable companies to automatically scan their networks for hardware and software and tabulate what exists and where, he says. Transition Partners has used these tools to help clients quickly and effectively determine what they should do with hardware assets.
In addition to asset management software, insurers can take other steps to improve decision making with regard to hardware. For example, Pettibone recommends establishing an effective IT governance process "wherein top management becomes equipped to make decisions regarding the balance of regulatory compliance with IT cost and timing," he says.
Effective governance will become even more important as regulatory compliance issues continue to add complexity.
"Over the next year or so, insurance companies will experience a great deal of change," Pettibone says. "It's clear that legislation will drive them to new and different products, markets and customers-and margins will be much smaller. Plus, regulatory compliance will be much more rigorous."
Insurance company CIOs "should establish basic architectures that will enable them to be very flexible and nimble," Pettibone adds. "They will need to rapidly introduce new products and services plus convert the existing base of business to new rules and regulations while maintaining today's high level of service."
Cross-functional business alignment is critical to making decisions about hardware, adds Gene Chao, SVP of strategic services, Americas Region, at Dimension Data, a New York-based IT services firm.
"The balance of function, cost and benefit [of hardware] has not drastically changed but priorities certainly do," Chao says. "Every CFO and CIO is balancing business improvement, automation, information availability/computing power, and cost of operations. The more clarity our clients give to us in terms of priority and 'must haves', the better solutions we can drive."
The Outsourcing Option
Some insurers have opted to outsource all or part of their IT infrastructure, which takes away much of the worry about if and when to replace hardware such as servers.
Privilege Underwriters Reciprocal Exchange (PURE), White Plains, N.Y., opted to outsource specific non-competitive and/or non-strategic technology components of its technological infrastructure, says Stuart Tainsky, SVP and CIO. This includes all the maintenance of its servers and networking equipment.
Outsourcing "has allowed the business to grow without a significant increase in overhead," Tainsky says. It has also given the firm the ability to keep its internal IT staff focused on high-priority, high-impact projects, he says.
By using service providers, PURE does not need to make decisions about upgrading or retiring servers. "It has truly taken out the need for us to worry about it," Tainsky says. "We work with our partners with specific SLA's on system response times and availability. If they need to boost the horsepower to get there, it is [they] who are responsible for incurring those additional costs. Outsourcing works for PURE because it eliminates those spikes in capital expenditures required to support a growing company."
Visit InsuranceNetworking.com to comment.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access