s a big fan of atmosphere and thematic environment, I queued up the "Blade Runner" Soundtrack on my computer. Vangelis' ghostly ambience, like the subject of this editorial, is futuristic in nature, but more importantly, the source of my audio bliss is the MP3 file format (produced from my own purchased copy of the CD, if anyone from the Recording Industry Association of America (RIAA)
Peer-to-peer (P2P) technologythe advancement which vilified MP3s and the concept of digital file sharinghas created a hubbub over the past couple of years, but the legal thunderstorm that ensued rained on the technology parade. Bring up the phrase "peer-to-peer," and the conversation undoubtedly will turn to copyright law, the Digital Millenium Copyright Act (DMCA), and the demise of Napster. But this isn't the fault of the RIAA, the Motion Picture Association (MPAA), or any specific lawsuit or legislation. The culpability is oursthe tech communityfor not seeing the forest for the trees.
Trade magazines and mainstream technical media have obsessed over file-sharing and its legally questionable future. Over the past couple of years, we have witnessed the market potential of a truly disruptive technology, but have halted its momentum by not seizing the real pearl in the Napster oyster: a change of mindset. Put simply, file sharing is not the most exciting and compelling use for P2P. File sharing has some valuable uses, but its primary purpose has been a prototype and proof-of-concept, one that has opened a doorway to something greater. P2P will turn computing upside down, as the next major shift in software architecture, as soon as we open our eyes to its real value and potential.
|Picture the enterprise, with 10,000+ computers, all of which sit idle 90 percent of the time. It is shockingly wasteful, yet no one seems shocked. I propose a simple change in the way individuals and enterprises think about hardware acquisitions: Every new computer purchase should add its resources to the existing heap.|
True Hardware Resource Utilization
Depending on the nature of the project I am working on, my home network may comprise up to seven computers. More than once while working through the quiet of the late night hours I have listened as one of my computer's hard drives is flogged with data, its processor light pegged solid as if painted. Meanwhile, my other six computers enjoy a state of restful hibernation.
It occurred to me one night, as I warmed my hands by the one slaving computer, the absurdity of our accepted norms for hardware purchases and operation. Every time I buy a computer, I have a new processor, a new chunk of memory, and a new hard drive, all of which are imprisoned in the cell of that computer case. Then upon running some resource-intensive application, the machine gets crushed with the processing load, while several other machines in close vicinity sit idle.
Or likewise, if I want to save a large file, I have to be sure that that particular machine's hard drive is not full, regardless of whether every other machine has 30GB of available space. Or similarly, finding my computer is running out of memory, while the other machines have plenty available. Each scenario is absolutely ridiculous. It is analogous to inviting your friends over to your house to help you move that grand piano in your living room, then giving everyone a crack at it, one at a time. The strength lies in the cooperative: the whole is greater than the sum of the parts.
Computers have three primary resources: processing power, memory, and storage space. Presently, we buy new hardware because a particular processing instance (computer) falls short, not because we are necessarily short on resources.
I should be able to leverage the entire union of resources at my disposal. And that's just me, and my puny processing requirements. Picture the enterprise, with 10,000+ computers, all of which sit idle 90 percent of the time. It is shockingly wasteful, yet no one seems shocked. I propose a simple change in the way individuals and enterprises think about hardware acquisitions: Every new computer purchase should add its resources to the existing heap.
P2P technology can make this possible, erasing physical boundaries and combining computing resource into one virtual resource heap that is utilized by any application on any computer within the collective. This would allow low-cost, older technology to be leveraged indefinitely, and drive computing costs down considerably. Imagine never retiring a computer, even when the resource requirements of the applications it runs surpass the maximum amount of memory or processing power the computer can support.