Touted as one of the hot new technologies in recent times, P2P technology has been not been quick to take off. Deb Mukherjee analyses the reasons behind this failure and areas in which it can evolve

Computers have now become ubiquitous. They are everywhere and are much more powerful than the ones we had a couple of years back. It’s an awesome phenomenon, but such is the ubiquity that we are not awed by it anymore.

Current day PCs, are more powerful than the servers of yore. They are more affordable and broadband has become more readily available. Concurr-ently, Internet technologies have drastically improved the increasing availability of broadband. This, together with more powerful PCs, has led to new business uses for Internet technology (intranets, extranets, using browsers to gain access to mainframe data, etc).

In early 2000, file-sharing, peer-to-peer networks such as Napster became extremely popular. Napster, allowed users to utilise the worldwide reach of the Internet to share music files stored on local hard disks. Although it was remembered more for the plethora of opportunities it offered for piracy, it gave to the world a powerful new platform for distribution. This led to a swamp of start-up companies surfacing, geared towards yet another paradigm shift called Peer-to-Peer (P2P) computing.

P2P technology extends computing to the edges of the network (such as Internet, intranet, LAN and WAN) by allowing different computers on the network to act as edge servers, thereby unlocking idle end-user computer resources such as PCs. The day is not far when P2P may prove to be a more cost effective, personally efficient, flexible and easily adaptable mode of person-to-person communication. It is already happening, and very soon the world will see it too.

For the uninitiated, P2P architectures enable computers to exchange information and resources directly between one another. A wide range of applications, ranging across multiple business domains, can embrace P2P technology for quick, easy and cost-effective solutions. Enterprises can leverage idle computing resources by enab-ling a direct exchange of services between computers that include exchanging data, shared processing cycles, cache storage and disk storage.

Does P2P have any business applications?
Not really. Not yet! In general though, business applications of P2P may broadly be classified into two categories data sharing and resource sharing applications. Data sharing applications, again, include file sharing, collaboration, intelligent agents and instant messaging. Resource sharing or distributed computing applications (applications that distribute compute-intensive applications across different resources within the network) include system resources such as CPU, cache, memory and disk.

File sharing in the P2P scenario involves protected content distribution across the network. Network traffic is decreased and thereby reduces the server storage needs by storing the data locally. Collaboration involves secure file sharing in ad hoc groups, while increasing productivity by decreasing the time for multiple reviews by project participants and allowing teams in different geographic areas to work together. P2P technology also helps utilise idle resources within a network such as CPUs, disk space and MIPS, thereby allowing businesses to distribute large computational jobs across multiple computers thus reducing computational costs.

P2P Platforms and Tools

This past year has seen a large number of companies building P2P tools and platforms. There are known to be somewhere around 120 start-ups already touting to be P2P players!

Today, we can successfully apply P2P computing architectures across multiple business domains that include financial services, biotech, CAD, film animation and search engines. Financial Services firms are emerging as frontrunners in implementing P2P architectures for their business solutions.

P2P enabled distributed computing offers financial firms a potentially efficient and cost effective solution along with speed, accuracy, protection, reliability and flexibility. In a business where time is money, this enables fast and instant access to important information.

One such approach that has proved successful, is the use of hybrid peer-to-peer networks within the enterprise. This type of distributed computing, that resembles traditional client-server hub-and-spoke architecture, uses the server to manage processing jobs, but relies on high performance desktop PCs to provide processing power. The proposed hybrid architecture is scalable and can be customised to meet the varying needs of individual customer’s. Implementing this architecture amongst the PCs in an enterprise minimises security risks from outside the firewall.

Financial institutions that implement this solution can take on substantially more business, make more timely decisions and take advantage of discrepancies in the market and better service its clients.

The Future

According to industry exp-erts and analysts, by 2005, at least 30 percent of large companies will use P2P computing. I share that view.

P2P has gained convincing popularity and the need of the hour is for users and enterprises to fully leverage their IT investments in this technology. The profits from this are promising, but what it remains to be seen if it can deliver on those promises.

The author is CTO, Cognizant Technology Solutions