Managed recovery services that protect people, systems and data
A team from Sungard AS configured the new multi-site environment, with primary infrastructure hosted at the Sungard AS Philadelphia data centre and recovery infrastructure at its Aurora, Colorado data centre. Sungard AS managed the data migration to the new environment and now provides a dedicated network to enable data replication between the sites, capturing all changes to Gerber's Oracle database in two geographically dispersed storage arrays.
As part of the initial deployment, GAVS and Sungard AS worked closely to tune and test the environment and achieve the service level agreements (SLAs) committed to Gerber: a 60-minute recovery point objective (RPO) for data and virtual machines for server recovery.
For ongoing management, GAVS applies its expertise in AI-based operations and predictive analytics with a suite of proprietary tools that closely monitor system health across many metrics. The tools and expertise are central to the GAVS Zero Incident Framework, which aims to detect incidents before they occur and apply remedies with minimal manual intervention. For Gerber, this approach translates into a highly stable environment for its business-critical applications.
The Sungard AS team supports GAVS and Gerber with its own proactive approach to infrastructure management, monitoring and reporting on key indicators including CPU and RAM use, network bandwidth, file system and capacity. The Sungard AS team reviews all metrics in weekly check-in meetings and monthly service update meetings with GAVS. When values fall outside of specified ranges, Sungard AS and GAVS collaborate to diagnose and resolve the issue before it can impact Gerber's operations.
Solving an intermittent issue; preparing for the future
Shortly after Gerber's move to the new virtualized environment, an intermittent issue occurred, affecting the replication of data in the Oracle database. The issue was troublesome because it caused a lag in data synchronization across the two data centres and had the potential to impact Gerber's SLA for a 60-minute RPO. Sungard AS assembled its incident management team and worked with GAVS to analyze root causes and come up with a solution.
"We looked initially at network and other possible infrastructure causes, and then discovered the issue: usage patterns in Gerber's business units that generated very high peaks in the data change rate," says Rajendran D, Client Systems Principal, Sungard AS. "These sometimes exceeded the capacity of our VMware vSphere Site Recovery Manager (SRM) replication environment."
The addition of a dedicated server for replication has resolved the intermittent issue for now, but Gerber's continued business growth will soon push the solution to its operational limits, necessitating a move to a higher-capacity replication solution. To prepare for that day, GAVS and Sungard AS tested an alternate data replication approach based on AWS Cloud Endure. That solution, already packaged as a standard Sungard AS offering known as Cloud Recovery – AWS, has ample headroom to handle Gerber's business growth and potential future acquisitions – and easily achieve Gerber's specified RPO.