iSam Blue Offers Choice for IBM i High Availability
August 22, 2018 Alex Woodie
IBM i shops searching for a high availability solution have several options to choose from. One that they might not be familiar with is iSam Blue, a small Utah company that sells a solution called iSB-HA. According to iSam Blue’s president, the company is doing well by competing on affordability and service.
If you’ve followed the history of high availability software on IBM i and its predecessor systems, the name Robert Seal will sound familiar. Before founding iSam Blue, Seal was best known as the original developer for iTera’s data replication software, called Echo2 Seal and iTera parted ways, and today iTera HA is one of four IBM i-based HA offerings sold by Syncsort (formerly Vision Solutions), in addition to MIMIX, ODS/OMS, and Quick-EDD.
While Seal’s development chops are not in question, he actually did not develop the guts of iSam Blue’s current product (the company was the North American distributor for Trader’s Quick-EDD product for years before switching). Instead, Seal hashed out a white label deal with Bug Busters Software Engineering for RSF-HA. When HelpSystems acquired Bug Busters in early 2017, iSam Blue retained the rights to OEM the RSF-HA (now Robot HA) codebase as the foundation of iSB-HA.
Robot HA and iSB-HA ostensibly have the same underpinnings, and share about 80 percent of the same code, but the products differ in significant ways, according to Seal. The user interfaces are a little bit different, and iSB-HA has some audits that were not in the original RSF product. But the biggest difference is unquestionably the switch process. Seal says he was not satisfied with the limited options for conducting a switch from the production system to the backup system in the base product, so he developed his own.
The original switch was composed of instructions written into a copy book that was then compiled into CL. There were two main switch options supported by that process, but Seal wanted a third. “I wanted to give the user the capability to do an additional scenario,” he says. “So if you’re doing a test backup without affecting the primary, well you should be able to define exactly what you want it to be, and run it just as if it was a regular switch. And that’s what we’ve done.”
Besides the switch process, Seal is quite satisfied with the product as a whole. “The software itself is extremely easy to use, extremely reliable, and frankly the apply engine is the best apply engine that I’ve seen,” he says. “And the sync process, frankly I couldn’t improve it.”
Keeping objects in synch across two or more IBM i servers is a notoriously difficult task. Seal says iSB-HA’s synchronization process doesn’t throw up the types of object replication errors that many other HA products do. That’s primarily due to its automatic retry process and the fact that the apply process itself doesn’t cause locks.
“That’s one of the reasons we replace a lot of the other HA products,” he says. “They just have to resynch a lot of objects and we don’t.”
When it comes to replicated data types and objects and support for various fail-over architectures, iSB-HA has its bases covered. “We handle daisy chain, one-to-one, one-to-many. And we do bi-directional,” Seal says. “Technically, we handle many-to-one, although that one never made any sense to me.”
Seal is putting the finishing touches on a new process in iSB-HA that will eliminate the need to conduct a traditional switch entirely. According to Seal, the process will automatically re-direct application requests to a working IBM i server if another goes down.
The process currently doesn’t work in setups that involve advanced database operations, such as triggers or constraints, he says. But it does bring the potential to greatly simplify failovers for certain types of workloads, such as Web applications.
“It’s mainly for subset of companies that have some amount of data that really needs to be up 24/7,” Seal says. “They don’t want any downtime. They can’t wait a half-hour or 15 minutes or five minutes for a switch.”
Seal is still looking for a good name for the new capability. Right now, I just call it the switch-less switch, because there is no switch for the switch,” he says. “You just point the user where you want to point them, and we handle the rest. To me, it’s the holy grail of HA.”
Besides the switchless switch, the company has other plans to enhance iSB-HA too, including new monitoring capabilities and eventually a GUI to replace the 5250 green screen interface that the product currently uses.
What you won’t see from iSam Blue is a single GUI screen for a certain process. “I want the GUI to be able to do everything we do on a green screen,” Seal says. “I don’t want the customer to have to switch back and forth. ‘Oh we have a problem so now I gotta get a green screen to fix it.’ That doesn’t make any sense. If you’re going to do it, do it right.”
iSam Blue is one of the smaller HA companies on the market. The size of its installed base, which measures in the double-digits, would probably be a rounding error on Syncsort’s customer spreadsheet. But that’s not stopping Seal from plying the trade that he knows best – and taking a few converts along the way.
The majority of iSam Blue’s software deals are replacements, with iTera and MIMIX as the most common targets, Seal says. The company struggled a bit in 2017 following HelpSystems acquisition of Bug Busters, but that business has come roaring back in 2018. “This year it’s really picked up a lot,” he says. “And it looks to be a banner year, actually.”
While the IBM i HA market isn’t as exciting as it once was, thanks in large part to Syncsort’s efforts to consolidate the market, vendors are still quite competitive to win your business. According to Seal and one iSB-HA customer that IT Jungle spoke with (who will be featured in an upcoming article), iSam Blue’s willingness to work closely with customer to get the job done is one of its big selling points.
“When somebody needs something special, we create it for them,” Seal says. “We have to do more and be more aggressive than anybody else, because we’re smaller than everybody else.”
RELATED STORIES
Logical Replication: Still the Mainstay for IBM i HA
Understanding IBM i Options For High Availability
