Case Study: ZS Associates
ZS Associates is one of the world’s largest business service firms, focused on delivering value in business intelligence using their Global Delivery model for operating and managing technology. The company needed an enterprise DR solution that could keep pace with their rapid growth. With its customer base expanding, the company could not risk losing three days’ worth of data by relying on traditional DR methods.
Data Center Environment
ZS Associates has two data centers, based in Chicago and New Jersey. These centers comprise of 700+ Windows, Unix andLinux servers, of which 95% are part of their private cloud and the remaining 5% are hosted in the public cloud. ZS Associates’ business intelligence hosted services are based on a unique combination of QlikView, Cognos, MicroStrategy, Tableau and Business Objects.Challenges
The DR test was for a complex, multi-tiered application that was going to be failed over as part of the test. This was a customer facing, revenue generating, production application of over 120 virtual machines (VM’s) running on VMware’s vCloud Air in their Texas datacenter which needed to be replicated and synced to a similar data center in Nevada. These VMs included working production, mission-critical systems of multiple application types and multiple customer SLAs.
ZS Associates used a combination of tape backup and image archiving solutions to protect physical work loads in their Chicago data center. On a periodic basis, the archives were moved to the New Jersey data center, acting as an off-premises backup location. In the event of an outage in the Chicago data center, the image archives would be restored on physical servers manually, yielding an estimated recovery time objective (RTO) and recovery point objective (RPO) of 72 hours. ZS associates required a solution that would:
1. Work with a bare-metal to bare-metal physical DR Architecture.
2. Utilize their existing network between the two data centers.
3. Easily perform testing without affecting the production servers.
4. Be cost-effective and simple to set up and maintain. The Result - DR with Audible Proof for Multi-Tier Production Applications
The Solution – Disaster Recovery with RMM
The Rackware Management Module (RMM) platform offers a complete disaster recovery solution with support for both cloud environments as well as bare-metal physical infrastructure.
Advanced features like selective sync allowed ZS Associates to create policies that could sync higher priority drives, directories and files to improve RPO and tune WAN bandwidth usage.
RMM also offered them the flexibility to customize RPO, RTO, and cost objectives through various pre-provisioned or dynamically provisioned models. Dynamic Provisioning significantly lowers costs by provisioning servers at the DR event as opposed to pre-provisioning, thus using no compute resources until failover.
Despite the advanced functionality, RMM was also able to make DR testing quick and easy by automating a large part of the process without affecting production servers. This led to a large reduction in the IT manpower needed to perform regular DR tests.
Working with RackWare, ZS Associates were able to:
• Cut RPO 83% from 72 hours to 12 hours
• Reduce RTO 67% from 72 hours to 24 hours.
• Reduce the number of IT resources required to manage DR and testing of a complex environment down to one FTE.
Additionally, RackWare gave ZS Associates the flexibility to reduce RPO/RTO even further by pre-provisioning select mission-critical workloads, while saving costs on lower priority workloads by increasing RPO/RTO through dynamic provisioning at a DR event.
What Did ZS Associates Say?
Key reasons Zebra picked RackWare include:
• A simple to use platform with a high level of automation.
• A solution that worked out of the box, requiring a minimum amount of customized professional services.
• An advanced technology that would be able to customize RPO & RTO in the future if business needs changed.
• RackWare’s understanding of business-critical environments and experience working with large data centers.
Mark Kocour, Associate Principal, Global IT: “We easily implemented and improved our disaster recovery strategy without buying additional hardware. We can now protect a larger number of workloads across our data center while increasing their availability. RackWare allows us to reduce downtime during an outage, with failover capabilities of complex and expensive high-availability solutions, but at a fraction of the cost.”
To learn more about how RackWare can help you protect your workloads in the cloud, contact us.