If you’re coming to VMWorld in Barcelona, be sure to take in some of the architectural delights of Antoni Gaudí. Barcelona hosts many of his works, the likes of which you can’t see anywhere else in the world. Park Güell, Casa Batlló, Casa Milà, and of course the wonderful Sagrada Familia… These are works of incredible beauty and striking complexity, not to be missed.

Gaudi’s towers and steeples are innovative in the way they stand with no obvious means of support. Columns lean at organic angles, and stone walls reach heights that baffled and amazed the people of his time. Remember, this was at the turn of the 19^{th} Century – There were no computers to calculate the distribution of weight, compression of materials, or ideal curves to support free-standing towers. How did he do it?

Gaudí used the catenary arch – The curve a hanging chain assumes when supported at its ends and acted upon by a uniform gravitational force. Based on this principle, he built what are called funicular models. He would hang a floor plan from the ceiling and attach lines to it so that the hanging lines would describe the towers and walls of a structure. These were scale models, and he attached bags of weight that were to scale for the weight of the stone he would use for the walls. When the model was done it gave him a wire-frame of the building, hanging upside down. He used photos to measure the curves, which gave him the calculations he needed.

If you visit the Sagrada Familia (above), you can see a funicular model on display in the basement gallery. It’s worth noting that subsequent computer models have confirmed the validity of Gaudi’s calculations for the Sagrada Familia – Work continues, and the project satisfies today’s civil engineering requirements.

The funicular model is a form of analog computer. It uses data points (the floor plan, height of towers, and weight of materials) in relation to dynamic forces (gravity acting on the catenary arch) to arrive at a solution. His use of materials in relation to the dynamics of gravity and weight solved problems that would require overwhelming calculations to solve with traditional measurements and mathematics.

To illustrate how well this works you can build a simple analog computer yourself, using nothing but a ruler and a wad of chewing gum. Place the gum at some spot near one end of the ruler, and smash it into an arbitrary shape. Now ask yourself, how would you use digital computing to calculate the balance point of this ruler? You would have to calculate the mass of the chewing gum, the volume of the arbitrary shape, and use that to derive the center of its weight – then calculate the effect of that weight on the balance point of the ruler! This is the brute-force way of tracking every metric to calculate the current state of the system.

Now stick out your index fingers, with about as much space between them as the length of the ruler. Place the ruler on your fingers, and slowly bring them together. As you do this, the weight of the ruler shifts from one finger to the other, and the ruler stays balanced on your fingers as they come closer together. Ultimately, your fingers will meet, with the ruler still balanced on them – You have used analog computing to find the balance point of this irregular distribution of weight.

What actually happened here? You:

- Defined your environment (the ruler with the gum)
- Chose the data points that matter (your two index fingers)
- Modeled the data in relation to dynamic forces (gravity, keeping the ruler on your fingers)
- Moved your fingers toward the desired state (the balance point), and let the dynamics of your system keep things in order

For anybody who knows VMTurbo, this should sound very familiar.

One question people ask is, how can VMTurbo solve intelligent workload management without capturing and storing exhaustive and overwhelming amounts of data? This gets at the heart of VMTurbo’s innovative approach. VMTurbo uses the dynamic forces of supply and demand. Each entity in the environment is an agent that buys and sells resources. An entity only needs to know about the resources involved in its own transactions – The resources it buys and sells. As entities shop for the best prices, the environment always tends toward the desired state.

VMTurbo Operations Manager is not an analog computer, but it does use a similar notion of modeling the critical data points in relation to dynamic forces, and modifying the environment based on those relationships. This is more efficient than brute force collection of every available metric from every source, and crunching numbers as though you’re solving a big-data problem. It enables VMTurbo to manage environments with thousands of sockets in a single appliance, keeping the data in memory, and responding to conditions in real time.

Barcelona is a city known for innovation and elegant design. As you visit the city, be sure not to miss the works of a timeless master – Antoni Gaudí. And while you’re here, be sure to visit the VMTurbo booth (B111). Nobody wants to say VMTurbo is the Gaudí of software, but we like to think we have an innovative and elegant solution.