Berry Bros. & Rudd is Britain's oldest wine and spirit merchant, having traded from the same shop for over 300 years. Berrys first supplied wine to the British Royal Family during the reign of King George III and has continued to do so to the present day.

Despite Berrys' history and the continuation in the belief in traditional standards of service and quality, it is also extremely innovative. In 1994 it launched one of the first online wine shops, www.bbr.com, which is now regarded as Britain's leading wine website and has been awarded Website of the Year six times in the last seven years at the prestigious International Wine Challenge.

The Berry's IT Team were under pressure to significantly reduce their server estate as they were running out of physical space and power at their offices.

Many tapes held offsite resulted in significant expense in both management effort and storage costs. A simplification of the backup process utilising snapshot technology was therefore required as this would deliver a significant reduction in tape management costs.

In addition the current disaster recovery plan required re-building systems and restoring data from tape. This would give Berry's a Recovery Time Objective measured in days and a Recovery Point Objective based on the last successful backup.

The recovery process required an enormous number of manual steps and was not easy to test on a large scale. A repeatable, easy to manage disaster recovery solution was therefore required that would deliver a significantly lower Recovery Point Objective and Recovery Time Objective.

• A high performance hypervisor with the highest consolidation ratios in the industry
• Centralised management for ease of use and rapid provisioning
• High availability features such as VMotion, HA and DRS
• An industry proven platform delivering "mainframe" reliability

EMC Celerra array was chosen based on the following:

• Availability - a virtual infrastructure is only as reliable as its storage platform
• Scalability - it needs to support Berry's needs today and well into the future
• Connectivity options - iSCSI, NAS and FC provide maximum flexibility
• Advanced features - snapshots, clones & replication for backup and disaster recovery
• VMware API support - vCenter Site Recovery Manager and Multipathing

SNS services were chosen based on their ability to provide:

• A proven track record in the field of virtualisation and disaster recovery
• VMware and EMC pre-sales expertise
• A detailed understanding of Berry's needs
• Delivery of the most appropriate and cost-effective solution
• Certified and experienced VMware and EMC engineers
• A documented project plan that keeps Berry in control
• Detailed deployment documentation to aid Berry in troubleshooting
• Post deployment training to enable Berry to support the infrastructure

The key objectives for the project were as follows:

• Implement three EMC Celerra arrays across three sites
• Implement asynchronous replication over IP between the EMC Celerra arrays
• Implement array based snapshots with backup to tape performed once a month
• Implement a process to recover a file, virtual disk, virtual machine or complete site
• Implement a VMware Infrastructure at each site with VMotion, DRS and HA
• Migrate the existing physical servers to virtual machines
• Provide comprehensive knowledge transfer and documentation