Read our other posts on this subject here and here. For a quick best practices overview check out this short video: Data Retention Best Practices for MSPs

Every time I have a build or buy appliance discussion with someone, I hear the Byrds singing an old Bob Seeger tune. It is like a soundtrack running in my head repeating the mantra, there is a season, turn, turn, turn… When I started in our industry it was a season for building computer hardware. That season has passed. Allow me to explain.

There are several differences between name brand manufactured computers and white-label systems.

Quality of Components

There is generally a wide gap between name brand hardware and white-label hardware components. Even name brand hardware has different classes of components. These are typically consumer, business, and server class systems. The criticality of the application dictates the class of hardware that should be used. That is why server class hardware is built to be so robust. I don’t know anyone who would try to run a server for production on consumer class hardware, that is just asking for trouble.

Within the classes of hardware we also see huge differences in component quality between name brand systems and white-label ones. What one manufacture calls business class may only be considered consumer grade by another, and most often, it is the white-label builders that are using the lower grade components to keep costs down. That lower cost can comes with a higher failure rate of these components.

System Builds

Let’s say the name brand manufacturer and the white-label manufacturers are using the exact, identical parts. We would expect to see the same failure rates right? You might think so, but that isn’t the case. The next important step in producing a system is the build, how the parts are put together. Are the systems assembled in a clean, temperature controlled environment? Are the build steps documented? Are the people assembling the systems building 10 systems a week or 10,000? Improper handling of the components can drastically increase failure rate.


If there is a failure, what happens? What is the warranty? Do you have 24/7 support? Onsite, or are you required to ship the hardware for depo repair? Warranty is a very important part of the hardware being used. No matter how high quality the components and the build, items fail and are damaged. When this happens, how is your customer going to like not having a system in place, or the down time associated with a hardware swap?

Options in Hybrid cloud

When a software company is deploying an appliance based hybrid cloud solution, hardware used is a critical decision. Build or buy, and if buy, buy from which OEM? If you can’t build better, buy. Once you choose to buy, buy to match the systems criticality. I don’t know about you, but I think a DRaaS solution may be one of the MOST critical systems. Because of that I would make sure the hardware being supplied is high-end server class gear. Also because of the criticality I want to reduce the odds of failure as much as possible. So to that end I’m going to pick a high volume, name brand manufacturer . I also want to make sure I have a comprehensive warranty that provides onsite support. Does your DRaaS provider use hardware that meets these requirements?

Red flags are warnings. People use them to wave traffic around construction, we tie them on stuff sticking out of the back of our cars, and matadors wave them at bulls. I’m not sure how you use a red flag, but I use mine anytime I see white-label hardware.

Steve Noel