I’ve been hinting at, and alluding to a benchmark we (the day job) ran on a new product for a while. I took a month to rerun these tests, verifying everything. I wanted to make sure that we got this right.
Because these are big numbers.
Then we sat on it for another month.
Give us time to reflect, what will people’s reaction be?
We slowly leaked a few pointers to people. The reaction has been incredible.
Before I provide the link, let me ask you, do you play poker? Poker is a game of asymmetrical information. That which is known, versus that which you suspect of others, or that which you wish to hold close. The cards you will see below, one might assume would be our fastest possible result. I wouldn’t bet on that assumption. In fact, I’d bet against it.
Here’s the link (appropriately decorated with marketing bits, our test case, and a snarky name).
The results are actual measurements, under non-heroic conditions. The units used on this test are designed and built specifically for one customer and are used in their operations; as there are a number of design and engineering choices in this system which make tremendous sense in this users' context, but not necessarily in a more general context. Think of these as purpose-built custom high performance tightly coupled computing and storage systems.
This said, the architecture is one we can leverage to address more general problems in high performance tightly coupled computing and storage. And we can build these types of fast systems for others. Which is why out private cloud offering was created.