In October 2018, the Florida panhandle was hit by Hurricane Michael, a Category 5 storm that set a new record for the strongest hurricane on record to make landfall in the area. Florida State University (FSU) already understood the need for a long-term continuity of operations plan, and the devastation Michael wrought on the surrounding areas solidified how critical a disaster recovery (DR) service is to their overall operations.
Disaster recovery as a service (DRaaS) is the business world’s advanced solution to unexpected disruption. Whether the threat comes from natural disaster, technical failure, human error, or malicious action, the result is still unplanned downtime and the need for recovery.
DRaaS concepts explained
Two primary concepts drive a DRaaS solution: the amount of data loss and the amount of downtime. Key performance indicators are:
- Recovery Point Objective (RPO). This is your appetite or risk threshold for data loss. On a timeline, RPO extends backward from the moment of a disruptive event. Financial services organizations demand a low tolerance for lost transactions and requires a near-zero RPO. A small business in a less dynamic environment, however, might be able to absorb minutes or even hours of email correspondence being sent-then-lost and be satisfied with an option that secures a longer RPO.
- Recovery Time Objective (RTO). RTO extends forward from a disruptive event. This is the time it takes for your business to become operational again—with servers restored to the state they were in before the disaster happened. In a business setting, it considers: are your customers locked out of your downed website on the busiest shopping day of the year? Can your departments access the resources required to carry on operations?
These concepts are central to making decisions among the available approaches to mitigating disruption and driving your business or organization toward the solution that is most appropriate and cost-effective.
Florida State University creates peace of mind
For FSU, hurricane season and its threat to the Gulf Coast is a consistent reminder to stay focused on DR. With thousands of students and staff members relying on key learning systems and with sensitive databases containing student information and payment data, FSU needed to reduce RPO targets from hours to minutes and measure tiers of RTO targets in hours instead of days. It also meant building a service that would integrate end-to-end with all applications and scale to support operations now and into the future. Additionally, they had a requirement that the solution be built, tested, and launched on an accelerated timeline while optimizing costs.
FSU selected Enquizit, an AWS Advanced Technology Partner, to design and launch their DRaaS solution. Enquizit’s approach included using AWS artificial intelligence (AI) and machine learning (ML) for a smart and rapid buildout of the infrastructure and focusing on simplicity and automation for the environment to operate at a high state of readiness.
To help shape and evaluate potential solutions, FSU addressed a series of questions with Enquizit:
1. How costly, in terms of time and data, can we afford for recovery to be?
As a higher education institution with thousands of students and staff members depending on critical learning and payment systems, FSU could not afford much data loss or downtime. Enquizit’s solution delivered a near-zero data loss with less than one hour of RPO, a four-hour RTO for Tier-1 services, and an overall RTO of 12 hours.
2. How long will the project take?
Enquizit designed and launched the FSU solution in just eight months. The team used its technological depth and experience in implementing cloud migrations and utilized proprietary tools leveraging AI and ML to quickly and accurately develop the replication baseline.
3. How many necessary steps are contained in the recovery process?
A simple and short runbook is one indicator that the DR environment is operating at a high level of readiness. Enquizit’s approach to implementing DRaaS focuses on minimizing manual intervention and leveraging automation. These elements remove common obstacles to maintaining a high state of readiness, making it easy to run training sessions and handle recovery should disaster strike.
4. Can all of our applications work with the solution?
DRaaS is only as good as its flexibility and adaptability to the systems your organization relies on. For FSU’s DR solution, Enquizit used a hybrid model of replication patterns, including:
- Host replication using CloudEndure Disaster Recovery
- Oracle Exadata host platform with RAC replication to Amazon Elastic Compute Cloud (Amazon EC2) using Data Guard
- File replication using SoftNAS
This end-to-end model ensures that no system, technology, or platform (legacy or recent) is left behind during disaster recovery.
5. Once in place, how easy is it to operate, maintain, and scale the solution?
The time and effort needed to manage a DR solution throughout its lifecycle are can be simplified by using AWS, and since Enquizit uses a plug-and-play approach, where 80 percent of all code is agnostic between clients, your solution is simple to add on to down the road.
6. How cost efficient is the service?
Among different DRaaS providers, costs must be weighed against requirements and performance targets. FSU determined that using AWS and CloudEndure Disaster Recovery provided an effective balance of desired capabilities while keeping operational costs low.
Whether or not hurricanes are an immediate threat, organizations today need answers to their stakeholders’ questions of what happens in a worst-case scenario. Having a comprehensive DRaaS strategy provides the baseline benefit of a continuous autosave and also advantages realized across operations and security. In today’s fast-paced, customer-centric business environment, DRaaS is no longer a “nice-to-have,” but a “must-have” in order to preserve vital business functions and data should the worst happen.
Learn more by visiting the AWS Organizational Resiliency & Continuity Help Center or by contacting your dedicated AWS account team at aws-california-team@amazon.com.
