… that the day job is building atop our siCloud platform. In the article (definitely read it!) there is an great discussion about what the fundamental differences are between what Lucera is aiming for and what more traditional commodity cloud vendors are focused upon.
When it comes down to it, the difference is architecting for density of VMs in the commodity cloud versus architecting for performance and low latency in the performance cloud (Lucera’s). The design decisions are very different. In the commodity cloud, you don’t have hard latency requirements. You can overload single machines, as you have a pool of cycle and memory resources to allocate. Contention isn’t so much of an issue, just use onset of contention to spin up other instances, and load balance in numbers.
Think of it as the bunny rabbit approach to defensive operational tactics … reproduce rapidly, and don’t care so much about the issues of any one rabbit.
The performance cloud is a very different beast. This is where Lucera shines in financial markets, while our friends at Sabalcore shine in more general HPC cycles for hire. There are a few others that do the real HPC cycle for hire, R-Systems, and a few others. But what makes Lucera different than Amazon, Rackspace, etc. is the architecture for performance. That performance we encapsulated in the design. Its very hard to beat that performance, and you aren’t going to do that with a massive Amazon cloud.
Again, a good article. Well worth the read.