This has been bugging me for a long time, and we have to address this in every discussion we have.
You can’t build cost effective scale out systems on cheap-ass hardware designs. Its woefully inefficient, the cost blows up to achieve the type of performance we can achieve often with an order of magnitude fewer systems (hey … thats less acquisition cost, less TCO, less power/cooling, lower management strain, smaller footprint, tastes great, less filling, …)
The only way people recognize this is when they actually try it themselves. And they discover that cheap (as in acquisition cost) usually implies cheap (as in design/implementation). And this rate limits you on performance. And scale out. And scale up.
The problem for us is that this has become the message from many of the software storage groups. And software application groups. Just slap the software on a box and you’ll be alright. For non-resource intensive code, that is fine. For resource intensive bits, the code needs to be on well architected and implemented hardware.
Its this message that will get amplified (slap it on cheap hardware and off you go), and its this message that needs to be modulated to something like “yes, you can test it on this cheap hardware, but when you are ready to deploy, you need to build upon solid foundations”. This is why we built our appliances, to fix exactly these problems. Make it as efficient and performent as possible, in a smaller footprint, so you need to buy (often far) less of our kit to achieve the goals, and hey, its pre-configured to be maximally effective. And hey, the software? Its open source. A truck could run over all of the sourcing touch points, and you could still get service and support. Try that with the closed appliances and vendors.
Its no accident that we dominate the benchmarks we’ve run, with single boxes as compared to rackfulls of our competitors gear. Its not messaging, its reality.