Thursday, June 26, 2008

Compute Remotely, Assemble Locally

There's an obvious relationship between cloud computing, "over the top" applications, open networks, open devices, importance of application program interfaces.

The ability to compute, store data and execute code remotely means it is more affordable than ever for small developers and individuals to create applications that are immediately available to users anywhere. The existence of those applications "on the Web" makes the Web a more-powerful platform for bringing applications of any sort to market. That puts business pressure on walled garden business models of all sorts.

The existence of cloud computing also means software is becoming unbundled from hardware to a large extent. Not completely unbundled; not unbundled for every application or service. In fact, some apps require tight integration to execute with the greatest elegance. But the direction is more in the direction of how people use PCs than how they consume cable television.

The application explosion, built on open platforms and APIs, also means new applications can be built on the shoulders of existing apps and applets. Assembling apps begins to be a process akin to what one does with Legos, to oversimplify.

That also means apps more often are created globally, assembled locally. That has implications for browsers, networks and protocols. To assemble apps locally means a premium for rapid response. If assembled apps are to mimic the feel of locally-stored apps, response time is a crucial requirement. This requires more than big, fast pipes. It means browsers that are much faster than we have used in the past. It means a computing architecture that does not require so much traversing of wide area networks to grab app elements.

The issue is to answer a question: “How do I pair together one customer that’s CPU-intensive and another that’s IO-intensive and have the sum appear just like a single, well performing application?”.

There is lots of room for innovation here. And lots of distance to cover. But it's coming, even if most users only gradually are being exposed to use of remote and locally-assembled apps.

No comments:

Will AI Actually Boost Productivity and Consumer Demand? Maybe Not

A recent report by PwC suggests artificial intelligence will generate $15.7 trillion in economic impact to 2030. Most of us, reading, seein...