Our total downtime for the switchover to V5R4 went from 12 hours to 55 minutes.
CUSTOMER STORY
The Business Need
KSB Hospital is an 84-bed acute care facility located 90 miles west of Chicago, IL. KSB Hospital, along with its network of 8 off-site clinics, is a privately owned, non-profit hospital offering a variety of ancillary services and providing numerous programs that reflect a desire to promote the overall health and wellness of its patients.
In perhaps no other type of organization is minimizing system downtime more critical than in a hospital. Regardless of whether that downtime is planned or unplanned, with lives on the line 24/7, hospitals simply cannot tolerate lengthy disruptions in operational continuity.
Two years ago, KSB Hospital made a business decision to go to paperless patient charts, a transition resulting in a total reliance on its information systems for all patient charts and mission critical applications. With no paper record keeping to fall back on, the hospital’s scheduled maintenance downtimes – 12-minutes daily and one-hour weekly – pushed the limits of acceptability, while other IT maintenance practices, such as Initial Program Loads (IPL) – a recommended periodic rebooting of the IBM i which required hours of downtime – were simply deferred. The pain of downtime finally became too acute for KSB when its payroll application provider required an operating system upgrade from V5R2 to V5R4, which would necessitate the hospital bringing down its system for 12-14 hours.
“Two years ago it wasn’t a big deal if we were down for an hour for backups or 3 hours for an upgrade or even 12 hours, but with our move to paperless charts it is mandatory that we have our systems up 24-7,” says David Ginn, KSB Hospital’s Director of Information Systems. In fact, the Illinois Department of Public Health and Medicare mandate that electronic medical records be accessible to patients at all times.
We have mission critical systems, so I’ve never tried to put a price on either, but I must say that not having ER docs screaming in your ear because the system was down the night before is priceless.
The Decision
Ginn and his staff were already very familiar with Maxava HA suite, having used it on a trial basis two years earlier. Maxava Business Partner, IT Solutions Group, had loaded the Maxava HA Enterprise software onto KSB’s existing IBM i model 810 using logical partitioning (LPAR). This allowed the creation of a ‘logical’ machine within KSB’s existing server, so as to replicate the process of moving all of KSB’s data to another physical IBM i server.
The exercise had not only demonstrated Maxava HA’s effectiveness and the speed with which data could be moved to another physical machine, it also allowed the KSB staff to interact in real-world terms with Maxava’s help desk and support staff.
So once the decision was made to purchase a new IBM i model 525 with the V5R4 OS, Maxava HA Enterprise was installed,enabling the payroll application to be migrated from the V5R2system to the V5R4 system and the non-payroll applications to be replicated to V5R4.
“Every Thursday evening we scheduled a one hour downtime window anyway, so we took that hour to shut down the users, perform the Maxava HA role swap, test the applications and bring the users back online,” says Ginn. “Our total downtime for the switchover to V5R4 went from 12 hours to 55 minutes.”
The Outcome
Although the payroll application’s V5R4 upgrade requirement provided the initial impetus, Ginn credits the move to Maxava HA and the associated reductions in downtime for allowing KSB to stay current in every facet of information systems management. “We intend to apply our PTFs (IBM i Access for Windows Service Packs) and do our IPLs more frequently,” he says. “Any big 12-hour downtime event in the past is now closer to a 12-minute event. That’s how long it takes to swap over to our HA machine when we do our system upgrades or disk upgrades, and users don’t even know they’re on the backup.” “Maxava HA provides us with two significant benefits – disaster recovery capabilities and uptime,” says Ginn.