In Part 4, we laid the foundation by examining the critical differences between Backup and Disaster Recovery and analyzing the role of Scheduling and Automation in choosing a solution. In this installment, we’re taking the next step by evaluating the remaining factors.
You can review the case study here.
As usual, we'll apply these considerations to the case study, analysing how each solution measures up. Every element will help us get closer to the final question: Which solution best suits the client's needs?
So, let’s dive in and uncover what these next factors reveal!
RPO vs RTO Requirements
When considering the client’s case study, what RTO and RPO strategy do you think best meets their needs? How might their requirements for different systems influence the solution?
From the case study, the client has varying requirements for their three systems:
ERP Application:
- RTO: 1-2 hours.
- RPO: Not explicitly mentioned but can be inferred as low, given the critical nature of the system.
- Additional Recovery Needs: Flexibility to restore either to the cloud or on-premises.
Information System and Library System:
- More flexible RTO/RPO requirements, as these systems are less critical than the ERP.
These differences mean the solution needs to be adaptable, balancing stringent recovery metrics for the ERP with cost-effective approaches for the less critical systems.
Now, let’s compare AWS Elastic Disaster Recovery Service (DRS) and Veeam Backup and Replication Service to assess their suitability for meeting the client’s RPO/RTO requirement.
From an RPO/RTO perspective, DRS excels in providing ultra-low RPO and RTO, especially for critical systems due to continuous replication.
Physical vs Virtual Servers
Based on our case study, the client is managing a mix of virtualized and physical environments:
- The ERP application is hosted in a virtualized environment, which is critical for the business and requires a more robust disaster recovery solution.
- The information system and library system, however, are running on physical servers, which adds complexity when considering recovery strategies.
The client's mixed infrastructure means we need a tool that supports both physical and virtual environments.
So, let’s see how our two solutions compare:
For virtual environments, both DRS and Veeam offer comprehensive coverage. For physical servers, both DRS and Veeam support physical servers, but Veeam offers broader support for various physical server configurations and operating systems. Both solutions support physical and virtual servers, but Veeam has a slight edge in terms of flexibility and support for more diverse infrastructure.
On-Premises vs Cloud Restores
The client is exploring both on-premises and cloud recovery options for their systems, with a specific emphasis on restoring their ERP system, which is critical for their business operations.
For cloud-based recovery, AWS Elastic Disaster Recovery (DRS) offers a streamlined process with the potential to quickly spin up instances in the cloud. However, while DRS can restore to an on-premises environment, this approach adds significant complexity and cost.
The client would need to first initiate recovery in the cloud (spin up recovery instances in the cloud which are sized to match the on-premises servers), then initiate failback to the on-premises environment. This process would not only be costly (the client needs to incur an unnecessary cost of recovery instances in the cloud, plus hefty data transfer fees) but it would significantly increase the RTO for restoring to the on-premises environment. Furthermore, considerations around bandwidth, network setup, and the configuration of recovery infrastructure make the on-premises restore with DRS less attractive for this client.
Veeam, on the other hand, offers a more straightforward and cost-effective approach for on-premises recovery.
Veeam has a strong focus on both cloud and on-premises backup and recovery, with features designed for quick restoration to both environments. Its restore process is far more simplified, and it provides the flexibility to recover systems back to on-premises environments with minimal complexity. Additionally, Veeam offers tools to handle the nuances of restoring large data volumes, which can ease the recovery process from a cloud backup to on-premises hardware.
Additionally, Veeam Backup & Replication allows you to restore different workloads (VMs, Google VM instances, physical servers etc,) to Amazon EC2 instances.
And that’s a wrap for Part 5! We’ve tackled some heavy hitters—RTO and RPO, the unique dynamics of physical and virtual servers, and the ever-relevant debate of on-premises vs cloud restores. These are crucial factors that bring us closer to deciding which solution might best meet our client’s needs.
But the story doesn’t end here. In the final part of this series, we’ll zoom out to look at the bigger picture: AWS-native solutions vs third-party alternatives in the context of Veeam and AWS Elastic Disaster Recovery (DRS). It’s a showdown that will weigh the pros and cons of these two approaches to help us determine the ultimate recommendation for the client.
So, what’s your call so far? Are you team AWS-native or team third-party? Stick around—Part 6 is where everything comes together for the grand finale! Don’t miss it.
Top comments (0)