A Brief History


Peer-to-peer computing hit the front page headlines of technical Web sites in early 2000. It came on the scene as a new paradigm in network computing, a new technology for connecting people, and effectively utilizing untapped resources anywhere on the Internet and the Web.

P2P boasted of a network of equals, in which the traditional client/server partitioning of functionality and communication was replaced. The new paradigm of servents (a term formed from server + clients), partners in computing opportunity, was upon us. A number of factors unfolding in computing technology were driving this trend. P2P was changing the balance of power, and giving rise once again to the "power to the people" mantra of the 60s. Only this time, rather than protesting for personal individuality, the protest was for machine and silicon equality. A new generation of protesters took to the digital highways and soon caught the attention of the world.

The key factors changing the balance of power included the increased availability of inexpensive computing power, bandwidth, and storage. This, coupled with the explosion of content and subscribers on the Internet, started the next wave of the digital revolution.

The widespread adoption of Internet-based protocols enabled P2P champions to develop applications over a global network, a network never before seen in this light. Previous impediments to connectivity had been removed, and what shone through was a wealth of computing power, with access to a massive amount of content.

Where the content resided was no longer an issue. It resided everywhere, within servers, within clients, within the fabric of the network itself. This liberal dissemination of data coupled with unlimited total access raised eyebrows.

Projections of digital asset ownership infringement and copyright abuse quickly became a reality. Users of Napster, the popular MP3 file sharing program, proliferated overnight. According to industry tracker Webnoize (www.webnoize.com), the Napster user base was estimated at 1.6 million users in February 2001. The exchange of MP3 files that month alone reached 2.7 billion, and control of content ownership was seen by the media as spinning out of control.

Needless to say, the U.S. courts blazed with a battle that resembled Samson versus Goliath. Here was a small dot-com startup taking on media giants and the corporate establishment.

Proponents of P2P lined up to proclaim that freedom and liberty itself were under attack. We were witnessing firsthand as technology disrupted the status quo and brought into question the very laws that govern us. The battle being fought in the courtroom would determine the future of our society, and more importantly, P2P technology!

And then the dust settled.

The court ruled against Napster, and it suspended file sharing in July 2001.

Back to the Future

So, does this close the chapter on P2P? Hardly. P2P is alive and well, and having a significant impact on the design and development of system architectures and commercial applications. It is having a profound impact on the Internet and the Web as we build the next generation of network-centric applications.

You will find that there is no universal agreement on what defines a P2P system. However, there are system characteristics that have surfaced, and a common nomenclature is being used to identify and describe P2P systems.

It has become common practice to express the relationship between two computing entities as something-2-something. For example, B2B is defined as business-to-business; B2C is defined as business-to-consumer; and A2A is defined as application-to- application the list continues to grow. These computing entities pose no problem of definition. Business is well understood, as is consumer and application; but what about peer?

The dictionary defines a peer as "a person who has equal standing with another or others, as in rank, class, or age." What stands out in this definition is the equality relationship: one person equal to another person in an important characteristic, such as rank, class, or age. Often these characteristics equate to the status or control an individual possesses within a community or society.

So what you extrapolate from this definition is the notion of equality between computing entities. Fundamental to P2P is node equality, in which a node is defined as any processing entity that exists as a particular and discrete unit. See Figure 1.1.

Figure 1.1. Peers represent liberated entities on a network of equals on which anyone can speak and listen.

graphics/01fig01.gif

If a person refers to characteristics such as rank, class, or age, what are the characteristics of a peer? What characteristics are we measuring to be equal?

The simple answer is functionality. Peers are assumed to possess an equal capability in functions or services that they provide. This is contrary to traditional client/server models, in which the server possesses greater functionality and control than the client.

Napster was not pure P2P. All nodes did not possess equal processing capabilities. For instance, MP3 files were indexed on a central server; only the files remained on and were accessed from peer components. We will have more to say about mixed models such as this in a moment.

What made Napster and other P2P applications interesting is the ease with which large networks of cooperating nodes could be assembled, and that these nodes lived on the edge of the network. The nodes were common PCs that dynamically assembled to form a distributed file system. The network was constantly changing. The nodes of the network were transient by nature. They would enter and leave the network at will. This is not the traditional approach to network organization, and is atypical of filesystems.



JavaT P2P Unleashed
JavaT P2P Unleashed
ISBN: N/A
EAN: N/A
Year: 2002
Pages: 209

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net