Allianz Life adopts tiered storage

Two years ago, David Kaercher, vice president of core services for Allianz Life Insurance Company of North America, had a dismal storage utilization rate of around 30 percent and little buy-in from business executives who didn't want to spend money on systems they didn't understand.

Over the past six months, Kaercher and his IT team have completed the first part of an ongoing project to create a tiered storage infrastructure. The move has boosted data utilization rates to 80 percent, aligned business and IT objectives and created a catalog of storage services that offers business units a straightforward list of fees for storage capacity, data backup, data archive and disaster recovery capabilities.

At a cost of US$150,000, the 150-day project has improved the relationship with the business side immeasurably, Kaercher said during a session at Storage Decisions.

"You can't continue to ask for millions of dollars each year without a reason. All they [the business executives] know is they give us a lot of money to do our job and they don't understand where that money goes. It creates trust issues," Kaercher said. "Now they understand where the money is going for storage, and they get an explanation they can use to benchmark us against external service providers."

Allianz Life, a US$14 billion, Minneapolis-based company with 2,500 employees, has about 400 IT staffers. Kaercher's department has 160 workers and oversees three data centers and 100TB of primary storage managed by three full-time employees. That storage supports 50 IBM AIX servers, five AS400 servers and 500 Intel-based servers.

In his approach to the project, Kaercher said there was a notable difference between creating a service catalog and penning service-level agreements (SLA) for the business. SLAs amount to a contract between IT and business units for guaranteed levels of performance and availability, while a service catalog provides business units with a consolidated menu of services and associated costs that they can choose from. Kaercher also performs service-level reviews on a quarterly basis to help manage business expectations.

Over the next two years, Kaercher's team plans to expand the project to create a virtualized infrastructure of pooled server and storage resources that can offer SLAs across end-to-end systems, dynamically allocating those resources on demand.

But the effort didn't start out smoothly. Kaercher said much of his initial strategy reflected where his department had been "and not where we're going."

"We had this very defeatist attitude. There was lots of hindsight but no foresight," he said. "There was no thought about where we were going to be in two or three years and how do we map our storage road map to the needs of the business."

Most of the business alignment and technology rollout was performed with internal resources, with about half of the US$150,000 spent on consulting from Glasshouse Technologies to drive documentation and processes.

To align storage services with business unit needs, Kaercher said he first had to evaluate the plans of six business units and their needs. Kaercher assigned "service delivery managers" from his IT department to define service levels for each business unit based on the applications they use. For instance, some applications, such as transactional databases, were considered Tier 1 and automatically needed to be replicated between two data centers 35 miles apart; other applications, such as print and file services, were Tier 2 and didn't require added redundancy.

Kaercher said Allianz's biggest challenges were around compliance with HIPAA regulations and the Sarbanes-Oxley Act, as well as encryption of customer information and disaster recovery.

Kaercher said "the light bulb went off" for his company that things had to change in data management when, in 2003, data backups where taking 23 to 24 hours. Kaercher first sold business executives on the need to update the company's IBM Tivoli Storage Manager software. Then, on the physical infrastructure side, Kaercher's team standardized hardware and consolidated nine disparate storage-area networks (SAN) into two and replaced a Fibre Channel switch infrastructure from Brocade Communications Systems with centralized director-class switches from Cisco Systems Network-attached storage from Network Appliance Inc. was deployed as a "second tier" for less critical data storage.

The IT team also developed a list of best practices for networks and storage.

Kaercher said standardizing systems and consolidating onto two SANs reduced the time needed for recovery of AS400 systems from 36 hours to six. "It's not a terribly difficult effort to go through. It's just a matter of what level of commitment you want to make toward communicating what you provide to the business," Kaercher said.

Join the newsletter!

Error: Please check your email address.

More about ACTAllianz Australia InsuranceBrocade CommunicationsBrocade Communications SystemsCiscoCiscoCommunications SystemsHISIBM AustraliaIntelNetAppTivoli

Show Comments

Market Place