How a legacy application can kill your Always-On Business dream

How a legacy application can kill your Always-On Business dream

The title of this blogpost may be a rough choice but it feeds my current thoughts on the current model of the Always-On Business.

Lately I have been meeting a lot of people around protecting their business data and helping them to achieve the Always-On Business. Something that I do notice is that a lot of people/companies have the need for (near-)continuous data protection but the applications which they need to have online all the time 24/7 are far from ready for their needs.

Years ago (before virtualization) applications usually ran on 1 or maybe 2 servers and the need for being online 24/7 wasn’t as big as now. Yet things have changed now and (mostly) thanks to virtualization achieving 24/7 is becoming a must.

While people moved from a server at the office to a (rented) server in a datacenter things drastically changed. Servers got redundant in terms of power, networking and storage. With virtualization most companies even moved away from local storage to some high end storage which is also redundant and provides the capacity for their needs.

Sadly virtualization also had an effect that people used P2V for their applications. This has lead to the fact that applications are outdated and aren’t ready for the new Always-On Business. Over time things have changed luckily and things like loadbalancers, clustering,… improved making it easy to achieve the Always-On business.

The biggest problem I find today is that people have the need for (near-)continuous data protection but they are still stuck with a legacy application and this is the biggest challenge: convincing them to move forward. While they moved on from physical to virtual to enjoy the benefits such as High Availability, DRS, vMotion,… the application within the virtual machine stood still.

I always ask them if there are plans for a short term migration since this will help them even better in the long run. If an application is high available (ready) and all the rest as well achieving the Always-On business becomes a breeze!

The idea I always propose is:
– You have high availability on the datacenter level thanks to multiple power feeds.
– You have high availability on the network level thanks to multiple switches.
– You have high availability on the virtualization level thanks to VMware, Hyper-V, KVM,…
– Change your legacy application to a new platform so that you have high availability thanks to clustering, loadbalancing,…

If you are working on the above scenario and manage to achieve this (or already achieved it) think about the next step! Add backups for fast recovery in case of data loss AND replication for fast recovery in case of a real disaster: you’re on the way to achieve the 24/7 Always-On Business.

Niels Engelen on GithubNiels Engelen on Twitter
Niels Engelen
Working as a Principal Analyst in Product Management for Veeam Software with an interest in anything virtual and cloud with a strong focus on AWS, Azure and Microsoft 365. He is also a VMware Certified Professional, a Veeam Certified Architect and gained the VMware vExpert award (2012-2022).

One thought on “How a legacy application can kill your Always-On Business dream

  1. Amen to this. It’s not enough to expect the Infrastructure to rescue you from everything bad in the world – you absolutely need to consider your software stack as part of your HA aspirations.

Comments are closed.

Comments are closed.