A leading Alternative Investment Fund Management (AIFM) company maintains a robust policy for managing potential large-scale disasters – in the event that its headquarters faces a catastrophic event, potentially destroying the building or the entire city, it must resume operations within three days. Periodic SLA reviews and tests indicated the need for a hardware update at the existing disaster recovery site, since the colocation lease was expiring. Additionally, the existing tape backup solution was cumbersome and prone to errors.
In an initial meeting, Yves Sandfort and Philip Kriener of comdivision learned from the infrastructure and end-user computing managers that the current disaster recovery setup, which was struggling to stay current on updates, was comprised of eight nodes plus dedicated legacy storage in a colocation space. The compute infrastructure was linked to the on-premises datacenter via two dedicated lines, with Veeam handling the critical application.
The existing backup and disaster recovery platform was both complex and costly. An imminent hardware refresh and, in the event of a disaster requiring a scale-up to maintain operations, the system was especially difficult to expand. A major disaster threatening the entire city's infrastructure could jeopardize operations at remote offices, including the Veeam backups stored in a tape library at the colocation, which further complicates remote recovery.
After a workshop with the customer to understand their needs, Yves Sandfort and Philip Kriener outlined three options:
The customer's representatives initially doubted their ability to justify the costs of a new solution to their board. However, as Sandfort and Kriener demonstrated during the workshop, the new solution was more cost-effective than the existing one – primarily due to the savings from avoiding a hardware refresh and reducing maintenance efforts by using VMware Cloud on AWS, as opposed to maintaining an off-site SDDC.
The breakthrough was the Veeam backup to Amazon S3, which would allow quick restoration of the SDDC in VMware Cloud on AWS in the case of a disaster.
The customer chose the warm-start solution due to its optimal cost-value ratio, and selected Frankfurt's AWS datacenter. Comdivision assisted in setting up the SDDC platform, including VPN tunnel, 2-factor authentication, and VM migration from the old disaster recovery site. Witness servers, an RDSH farm with Windows 10 desktops for emergency jump hosts, and Veeam backups populated with AWS Snowballs were established. Direct access to the S3 Buckets was prepared. "We can now re-establish production within hours by adding new hosts to the SDDC and comdivision's guidance was pivotal; sometimes, external insights and collaboration are necessary to develop a solution. It was crucial that Yves and his team involved us in the decision-making process," said the infrastructure admin, highlighting the preparedness to provide 90% of the 640 desktops to colleagues within 2-3 days.
