SafeData Launches First Fully Managed Service for i5/OS HA
September 9, 2008 Alex Woodie
Have you had enough trouble trying to keep up with your i OS high availability (HA) environment? Not 100 percent sure that you could successfully execute a role swap when the hurricane is bearing down? If this sounds like your situation, you may want to consider the HA management service that was recently announced by SafeData. As part of the program, SafeData will take over all aspects of managing your iTera HA or MIMIX environment, and for about $20,000 per year, guarantee that it will be ready for a role swap.
While improvements have been made to i5/OS HA software over the years, the fact remains that implementing and running HA software is not a trivial exercise. Users must pay close attention to the state of data and object synchronization to prevent the source and target machines from getting dangerously out of synch or suffering from other problems, thereby putting transactions at risk of being lost if the primary machine goes down.
In an industry that loves change and thrives on “the next big project,” HA and DR processes often suffer from neglect. Once an HA system has been installed, many organizations will simply leave it as is, under the false assumption that it’s providing a net of protection for their vital DB2/400 data and i5/OS applications. As many as 50 percent of i5/OS HA implementations are basically worthless in the event of a disaster, according to some estimates.
In other words, the HA safety net has a lot of holes, and it’s largely due to poor HA management practices.
Peter Briggs, president of SafeData, has enjoyed a first-row seat to the HA game. Before launching SafeData’s HA hosting service several years ago, he worked at an AS/400 reseller that implemented scores of HA environments. Today, SafeData provides hosting services for about 100 AS/400 shops–about 40 that subscribe to SafeData’s hosted HA offering, and about 60 using its DR vaulting offering.
Briggs has seen many an HA environment go south due to neglect. “Once they make an HA purchase and get it synched up, they leave it. Then they don’t maintain it properly,” Briggs said in an interview with IT Jungle. The reasons? Complexity and a lack of focus.
“There’s a lot of instability and lack of confidence in the solution unless it’s maintained properly,” Briggs said. “The reason you need to test it [and maintain it] is because it’s complex. It remains complex. We run it all day, every day, for all kinds of companies, and we see all kinds of things pop up.”
Over the years, several of SafeData’s customers inquired whether the company could do a little more for their HA environments. SafeData already offered some services–monitoring the network replication and maintaining the server. However, managing the HA environment remained the customers’ responsibility.
SafeData relented, and started managing the HA environment of one of its customers, although company officials believed at that time that it was a one-off situation. “Then more clients started asking us if we would run the solution for them, so we recognize that it was something customers would be inclined to look at,” Briggs said.
In late August SafeData officially announced its new Managed High Availability Services offering. Under the program, SafeData personnel will perform all activities necessary to maintain an HA environment, including monitoring replication, performing audits, testing role swaps, downloading updates, and anything else required to keep the HA environment in-synch and ready to be utilized. The company will also initiate and execute failovers in cooperation with its customers.
SafeData will manage any MIMIX or iTera HA environments; and the company is looking at supporting the DataMirror HA environment, which is now owned by IBM. It doesn’t matter whether the applications live on servers that are installed in SafeData’s data centers or on servers installed at the client’s location; the company’s personnel can manage the software remotely.
The cost ranges from $1,200 to $2,500 per month, depending on the complexity of the environment. The contract includes a service level agreement (SLA) that guarantees the HA environment will be switch ready. However, due to the complexity of environments and the fact that SafeData doesn’t run the business applications, the SLAs won’t dictate a minimum amount of time to conduct the switches.
Briggs thinks he has a good business plan, considering the thousands of borderline HA implementations around the world offering little or no protection. “My view of this thing is, if you’re going to spend $100,000 on software, to spend another $20,000 to make sure it’s run properly is a good value,” he said. “To do HA properly, it takes half a body, say $50,000 per year. We’ll do it for a lot less.”
Six or seven SafeData customers are already utilizing the fully managed service, and Briggs just hired two more techs to prepare the company for growth. As the business grows, SafeData plans to leverage its growing base of HA knowledge and drive more cost out of the equation.
But the cost will never go to zero. Despite all the “autonomics” and other whiz-bangs vendors build to simplify HA management, good HA practice still comes down to having boots on the ground.
“Once we get them up and running, it’s still not a perfect scenario,” Briggs said. “Even though we’re monitoring and maintaining it on a regular basis, there are still problems with COM lines dropping. Things get kind of hairy sometimes. We see it all the time. The difference is we correct those issues as they come up, instead of putting them off.”
In other news, SafeData yesterday announced enhancements to its DR services. Company officials claim, by implementing a virtual recovery solution at the target vaulting facility, full restoration becomes up to three times faster and gets customers back online within three to 10 hours.
“Disk-to-disk backup that is encrypted and copied to several different storage pools and combined with a guaranteed recovery in just hours is a compelling service offering for System i shops,” Briggs stated in a press release.
For more information on SafeData’s Managed High Availability Services or its System i Recovery Service offerings, visit the company’s Web site at www.safedata.net.