WTS Expands Into Managed DR for i OS Apps
June 9, 2009 Alex Woodie
JD Edwards outsourcing specialist WTS is expanding its managed disaster recovery (DR) services. As part of the new offering unveiled last week, WTS will provide the data center space, the System i hardware, and the human IT expertise needed to get a customer’s critical i OS (i5/OS) applications and data back online and functioning in the event of an outage. The company will support any i OS applications–not just JD Edwards–and offers several ways to transmit data, depending on specific recovery objectives.
WTS was founded in 1997 by Tom Hughes, an influential JD Edwards customer, to do one thing: be a JD Edwards hosting provider. Over the years, dozens of companies have signed contracts to have WTS run their World and EnterpriseOne ERP systems for them. While ownership of the prominent ERP system changed hands several times, WTS remained focused on its JD Edwards hosting business, and today it still considered the premier JD Edwards hosting provider by Oracle, which owns JD Edwards, not to mention 49 percent of WTS.
In recent months, WTS has started to explore other lines of business besides JD Edwards hosting. In March, it announced a new co-location business that allows JD Edwards customers to stash their servers at one of WTS data centers in Seattle, Washington, and Denver, Colorado, but to continue to manage and service their World or EnterpriseOne environment themselves. (In traditional outsourcing, the hosting company manages all aspects of the ERP system.) Still, WTS didn’t wander from the JD Edwards patch.
Then last week, WTS announced Managed Disaster Recovery, its first service that isn’t directly tied to JD Edwards. Under this new service, WTS is offering to run any i OS, i5/OS, or OS/400 application for a customer on its own Power Systems, System i, or iSeries hardware in its Seattle and Denver data centers. It doesn’t matter whether that customer environment includes Infor BPCS, Lawson S3, or Inovis EDI–WTS will figure out how to make it run, and recover the environment with the latest backup in the event of a disaster.
WTS espouses and end-to-end approach to managed DR that includes a higher level of customer service and human interaction than a customer will typically get by signing a contract with one of the big DR providers, such as IBM Business Continuity and Recovery Services (BCRS) and Sungard Availability Services.
“If you look at an IBM Business Continuity or SunGard solution, in that model the provider has given the customer access to a data center location and access to equipment upon the declaration of disaster, but the customer still has to perform the recovery,” says Ward Quarles, director of marketing and communications for WTS. “They still have to get people out of the current scenario and on an airplane to one of these sites across the U.S. where these recoveries occur.
“In our model, we’re eliminating that customer travel component. WTS will provide data center space or equipment, yet our staff will actually perform the recovery on behalf of our customer once a disaster declaration is issued. So we’ve eliminated that people and travel requirement that goes along with the traditional model.”
Eliminate the travel requirement, and you eliminate a single point of failure, says WTS chief technology officer Richard Dolewski, who is spearheading the company’s new managed DR offering.
“The single point of failure for most every IT infrastructure scenario is the underlying fact that most companies still hope that in an actual disaster, their staff will be available to travel to their recovery site and execute the recovery,” Dolewski says. “Whether companies have tape backup, vaulting, data replication or high availability based solutions–no matter if it’s an internal datacenter, contracted facility, quick-ship program, or any other combination–no staff means no recovery.”
WTS’ managed DR offers different levels of service, depending on the customer’s recovery time objective (RTO) and recovery point objective (RPO). If the customer’s core application requiring protection is a payroll system, and payroll is only run once every two weeks, then the managed DR mechanism may not need to change much from its current tape-based backup systems and offsite storage at Iron Mountain or a similar vendor.
However, if the customer is doing business on the Web 24/7 and needs to get critical systems back up and running as quickly as possible following an outage, then WTS may recommend real-time replication or high availability using a product such as Vision Solutions‘ MIMIX. Customers that fall in between and have moderate RPOs in the 12- to 24-hour range can probably get by with the network-based backup replication mechanism; WTS uses the Avamar backup and data vaulting software from EMC to move backup data from customer sites to SAN arrays running in the Seattle and Denver data centers.
The exact method of data transport and which applications to target for recovery are determined by WTS following an on-site assessment of the customer’s IT processes. “We’ll go to their site and do a backup analysis to make sure that what they identified as critical production data is actually on the media that they’re currently storing,” Quarles says. “Once we prove that all production ready data is there, we will actually do a test to prove that it can be recovered in our data center. That’s when we build our run books, or DR recovery plans. That way, if the customer had to declare a disaster, we’ve already been through the scenario a minimum of one time.”
WTS insists on testing its customers’ recovery procedures at least twice a year. All too often, organizations fail to test their DR procedures, exposing them to the risk that their DR procedures are ineffective and won’t work in the event of an actual disaster. WTS seeks to minimize that risk. “We are continually proving they can be recovered in our centers with our staff, with little or no interaction with their organizations,” Quarles says.
WTS isn’t new to the managed DR business. It has been offering DR as a service to its JD Edwards customers for years. However, this is the first time WTS will be supporting i OS applications other than JD Edwards. Will WTS employees be up to the task of supporting a diverse array of i OS applications, including older ERP systems, modified applications, printing checks or invoices, or running EDI transactions?
Yes, Quarles says. “Through the customer’s support of that initial testing process, our run books will define to that WTS employee what success looks like in the recovery of that specialized solution,” he says. “Customers validate after recovery that we have achieved that level of success. So it’s cooperative during the discovery phase, and reinforced for each of those testing periods.”
WTS has dozens of AS/400, iSeries, System i, and Power Systems servers, and runs many different versions of the OS/400, i5/OS, and i operating systems. However, WTS doesn’t run every OS release, so a customer may be out of luck finding support for an early release of OS/400 V4, for example.
In addition to i OS applications, WTS is offering to recover Windows applications running under SQL Server databases, and Unix or Linux applications running under Oracle databases; JD Edwards EnterpriseOne, after all, runs on many different platforms, and WTS is outfitted with thousands of xSeries servers to support these environment. It also supports Microsoft Exchange e-mail with its new managed DR service.
Another factor separating WTS from IBM BCRS and SunGard, Quarles says, is the fact that WTS does not charge a fee when a customer declares a disaster. What’s more, the customer is not charged any additional site access fees for the first 10 days. If a natural disaster requires the customer to rely on WTS for longer than 10 days, a long-term DR contracts or short-term hosting agreements can be arranged.
WTS is hosting a Webinar on its new managed HA offering tomorrow at 11:30 a.m. ET. Interested parties can sign up for the Webinar, which is titled “Is Your System i Recoverable?,” at the WTS Web site at www.wts.com.