Unisys ClearPath MCP: Capacity Management

"A major heath insurance company had taken over 2 other companies, each of almost the same size. They needed to save costs by merging their IT operations from the existing 5 to 2 data centres. They also needed to take their disaster recover capability from an expensive outsourced vendor to an in-house facility. The company had been using our products to ensure that they provided an excellent service over the previous 5 years. Their hardware vendor had proposed 2 massive identical solutions for production and backup facilities, each being over 4 times the size of their existing facility. We worked with the client to produce 2 much smaller configurations. These were modelled on running 3 copies of the existing configuration and a 4th copy for development and testing. In a disaster scenario, production could continue through loss on any one of the 4 partitions. In the unlikely event of a whole site being unavailable, the remaining site would be reconfigured as a single system that could run the combined work of all 3 servers with a temporary increase in processor power. In this way, the client was able to continue to deliver the same excellent service at 50% of the cost."

“A manufacturer of fitted household furniture was suffering severe overnight batch run overruns with their parts assembly and distribution processing. This was impacting their daytime online ordering and was having a detrimental effect on their highly competitive business. After reviewing their operation over a week, we identified severe limitations on their disk configuration. By regrouping their disks, we were able to remove this bottleneck. However, the resulting 20 percent reduction in run times was not sufficient. With the disk bottleneck removed, this exposed application and database issues. We increased both the cache memory on the database and the disk retrieval sizes. This decreased the run times by a further dramatic 60%. They were then able to support their new year’s sales rush for the first time in 3 years, resulting in increased profits. The following week, the client suffered an overnight batch failure, but, due to the reduced batch times, was able to reload their database and re-run their batch in time for the next online day.”