M&A in our space

The day job’s products have never been stronger, fit together as well, or had as great a story arc as they do today. We can deliver denser, faster, easier to setup and manage systems quite easily. Our application stacks run atop this system on our ample computing power, and we provide massive network pipes in/out, as data motion is hard. Many more cool things are coming, but for now, we are working very hard on building something awesome.

So are others in this space. And the wheels of M&A are running fast. One of our suppliers for a project was acquired weeks before delivering us a critical bit for one very awesome system, and we’ve been scrambling to replace them (as they lost interest in the project and projects beyond a very limited scope once they were acquired). Many of our flash suppliers have been acquired by the big strategic OEMs. In one case, fallout from that managed to derail a very important deal for us … and annoyed me beyond belief. The customer wound up with, what I believe to be, inferior gear, at an inflated price. I am not sure how their performance is, but we win our benchmarks by very decisive margins for a reason. You can’t fake performance. You can’t rely upon well known brand names to be able to perform at the level of purpose built hardware. You have to decide if brand names are more important than performance. If acquisition cost per server is more important than the performance goal.

I’ve noticed that about every 2 years or so, someone steps up to ask us if we are for sale, and then proceeds to make a mindbogglingly terrible offer. The last time was about 2 years ago, so we are due for something soon …

Valuations in our market have skyrocketed, so this may deter some. But as I noted, you can’t fake performance, you can’t fake density. If you really care about these things, you need to be working with the best. If you want to own that market, you need to make sure you have the best stuff to offer. Or you are going to get your clocks cleaned.

You can always try to copy designs, but that carries its own risks.

Should be an interesting year.

Viewed 44191 times by 5794 viewers

Facebooktwittergoogle_plusredditpinterestlinkedinmail