Software Reliability

Software Reliability

Software Reliability

Software reliability can measure many metrics. One metric is how often the software is available under heavy stress from users, which we address in the Scalability page. We choose to focus our reliability numbers on our ability to run successful auctions 100.00% of the time. There has never been an auction for one of our clients that we have not finished successfully for them. This is a testament to our software and our software team, who have worked hard to make our software the best in the world.

One of the huge advantages of using Optimal Auctions software compared to a smaller company, or even a start-up, is that you are taking advantage of all the auctions in the past 15 years that we have run for clients. The software we are running today is the same software we were running years ago (updated and improved of course). But, our clients have the advantage of running the same auction software that has completed dozens of auctions and sold hundreds of millions of dollars of product. This means that any software bugs were rooted out of the software years ago in past auctions. Problems in the auction designs were identified and fixed a long time ago. Usability problems with the bidding pages were seen and fixed long before your bidders will log in to the software. You are "standing on the shoulders" of all the past clients we've had, and get to take advantage of their past success for your own auctions.

The end result of using software that has been tested tens of thousands of times by our dozens of clients and thousands of bidders in over a hundred auctions is that all the major bugs in the software are gone. Now, of course, we can't claim that our software is bug-free, because no software is. But, we can confidently state that all of the major bugs and issues in the software are long gone, and any issues that may remain are so small as to be negligible to the proper functioning of the software.

We've talked with some clients who have considered going with a start-up to run their auctions - we can't recommend against this enough. A start-up may be cheaper and they may make big promises, but one thing just won't change - their code is brand new and will 100% without a doubt contain major bugs. It's a law in software development. New software contains major bugs. No level of testing can make a new product bug-free. Only time can fix all major bugs. When you're considering a start-up to provide auction software, you as a client have to ask yourself if you're willing to accept those major bugs in the software, and whether you want to put the success of your auction at risk.

Because we re-use our software for every auction we run, and have done so for over a decade, our product is reliable beyond fault, able to run auctions successfully 100% of the time.

Hardware Reliability

Hardware reliability is dependent on our server hardware provider, Datapipe. However, Optimal Auctions has chosen Datapipe for their award-winning reliability, and have trusted them with the hardware that underly our auctions for more than 8 years.

Hardware reliability is measured using a metric called "uptime" - basically, what percentage of the time is the server working properly and reachable by users. In terms of auction software, a server that is "up" can run the auction correctly, and a server that's "down" cannot run the auction correctly. Even the biggest and best websites in the world, sites like and can never quite hit 100.00%, because sometimes things happen. For example, every once in a while, is so overwhelmed by traffic for the newest product that their website doesn't respond. And Amazon spends billions of dollars a year ensuring their reliability!

Datapipe has excellent reliability for their cloud servers - 99.99% as rated by Netcraft, an independent server rating company. This is the 7th highest rating of all the many server management companies in the world.

To put 99.99% reliability into perspective, that is only 1 hour of downtime for an entire year. Only 1 hour out of the entire year, and most auctiions by our clients only run for 2-8 hours. The chances of that 1 hour of downtime happening during the few hours that your auction is running is infinitismally small. It's as close to 100% as you can get without offering a guarantee!

In addition to the overall network reliability, we also take advantage of the power of cloud computing. All of our auctions are now run on cloud servers, and this is a huge benefit over running auction on a dedicated server, because of the nature of the cloud. In cloud servers, if a physical machine fails, which happens about once a year, then the "cloud" infrastructure simply moves all the computing to another working machine. That's the beauty of the modern cloud. If you instead have a dedicated server, if that server fails, there's no chance of recovery and no fail-over in place.

To compare our hardware reliability to our competitors, many of our competitors choose to operate their own dedicated servers in their own office to host their auctions. In our opinion, this is an incredibly old-fashioned approach to software, a solution that belongs in 2007 rather than 2017. More importantly, it's incredibly risky to your auction. Incredibly risky!

Let's examine the math: it's estimated that 20% of dedicated servers will physically fail every 4 years (and we know our competitors aren't upgrading their servers that often). It's also estimated that in-office networks fail 6 times a year, on average for 2 hours each. That's 12 hours of down-time a year, with a 20% chance that the machine will simply fail in the middle of the auction. Compare that to our approach using cloud-based servers, with a 1 hour downtime yearly average and a 0% chance that the machine will simply fail. You can see the incredible risk that companies take when they choose one of our competitors who use their own dedicated servers.

Choosing Optimal Auctions means your auction is 12-20x more likely to succeed without any issues!