Seven minutes to full disaster recovery for the HP NonStop system at the Ontario Ministry of Health and Long-Term Care

Seven minutes flat. That’s all the time it took the Ontario Ministry of Health and Long-Term Care to switch its Health Network System (HNS) from its primary processing environment to backup disaster recovery procedures when a network outage occurred during peak production hours.

About the HNS system

The HNS system has relied on HP NonStop servers for more than 12 years and now provides the highest levels of availability for around-the-clock patient drug profile information used in hospital 24 x 7 emergency departments, for adjudication of drug claims for pharmacies, and for other related services. Because the HNS system is critical to the Ministry and to Ontario’s pharmacies and emergency services, HP NonStop Remote Database Facility (NonStop RDF) software was chosen for use in its disaster recovery implementation.

Operated and managed through the services of HP NonStop partners ACI Worldwide and Green Shield Canada, the HNS primary site is located in Windsor, Ontario, and houses a NonStop S74006 server, soon to be upgraded to a NonStop S88006 server. The backup site, located in Toronto, is designated for disaster recovery, batch processing, and hosting of a large historical database. It runs on a NonStop S74008 server, soon to become a NonStop S88008 server. The two sites are connected with redundant high-speed communication lines and rely on NonStop RDF software to synchronize primary and backup production databases comprising 50 gigabytes of data contained in 540 files.

Practice makes perfect

While NonStop RDF software takes mere seconds to enable a switchover to backup computing, it is the human and system processes and procedures that must be mastered in order to make a rapid change that enables full business continuity.

With the human element in mind, disaster recovery exercises are performed every three months for the HNS system. These exercises ensure that procedures—such as communication-line process startups, readiness validation, and other operational tasks—are continuously tested, automated, streamlined, and improved. Exercise sessions also ensure that staff training and switchover process awareness are fully maintained.

Putting NonStop RDF software to the test—in seven minutes flat

Recently, the HNS disaster recovery implementation was put to a real-world test during prime production hours. When a network outage at the primary Windsor data center seriously impacted HNS system availability, the Ministry of Health was confident in its decision to switch its online transaction processing (OLTP) to the Toronto backup site, knowing that NonStop RDF software was maintaining a completely current replicated database.

The results were flawless. Only seven minutes were required to bring up the backup site for full online processing. When network problems at the primary site were resolved the next day, processing was switched back without incident or delay. NonStop RDF software maintained full synchronization between the two databases and used nonaffected HP Expand paths for communications.

“This single event not only demonstrated the value of disaster recovery testing, but also validated the decision to rely on NonStop RDF software to safeguard the availability of the HNS system,” said Jim Moore, director of Managed Services for ACI Worldwide (Canada) Inc.

More about NonStop RDF software

A number of selection criteria were important to the Ontario Ministry of Health and Long-Term Care when it chose NonStop RDF software for the core of its disaster recovery implementation. NonStop RDF software offers

What sets NonStop servers apart are their unique, loosely coupled architecture, designed from the ground up to provide unprecedented reliability, manageability and scalability.

• Excellent product support and maintenance

• Easy configuration and management

• Leverage of existing network and system resources

– Eliminates the need to modify application code by using HP NonStop Transaction Monitoring Facility (NonStop TMF) logs for input

– Utilizes Expand software and existing network redundancy communications for intrasystem communications

• High throughput when updating backup databases for minimal delays

• Support for multiple target systems, an important feature during the Ministry’s NonStop K-series to S-series server migration

NonStop RDF software makes it possible to create a disaster tolerant environment that can share processing in real time across multiple systems or physical sites. The software replicates database changes across systems using high-speed, low-level operating system interfaces. The key to continuity or recovery after a disaster is having all data on another system before a failure, and nothing outperforms NonStop RDF software’s low-latency, high-performance database replication. Any system in the NonStop RDF network can take over the work of another system in a matter of seconds. There is no need for system reboots, application restarts, or database reconfigurations. Nor do local workloads on the target system have to be jettisoned. NonStop RDF software handles even the largest application configurations and can support multiple live sites backing up each other, with every system running live transactions.

Find out more today about business continuity and disaster recovery software for NonStop servers.

This news is published on NonStop Computing websites. December 2005.