Time for a bit of optimistic telecom news

By Johna Till Johnson

Network World, 08/19/02

This week marks a milestone for me: After 14 months of writing this column twice per month, the folks at Network World have asked me to take it weekly.

To mark the event, I'm going to take a break from the cumulative drumbeat of horrible news in the telecom sector. Yes, the news is still bad, but let's look instead at three developments I think will have a significant positive effect on how telecom services are delivered.

The first is grid computing. Basically, grid computing refers to linking arrays of low-end and midrange machines to take advantage of unused CPU cycles for processing-intensive tasks such as modeling and simulation. For commercial users, the big driver is cost savings. Instead of spending tens of millions of dollars on high-end supercomputers, users can achieve the same or superior results with off-the-shelf hardware (some enterprising companies are even building grid-computing arrays using obsolete PCs). Financial services firms are beginning to deploy grid-computing techniques for predictive financial modeling, and over the next 12 to 18 months you'll see more following suit.

The telecom impact is software and solutions developers are working on architectures that enable grid computing across the WAN. That will require a lot of quality of service because latency and jitter become critical, along with plain old bandwidth. But it's a great excuse for service providers to begin marketing "grid-optimized services."

The next big development is quality-of-experience (QoE)-enabled networks, spurred on by - get this - online gaming.

Back when I was a newly minted engineer at my first job, it dawned on me that gaming technology typically is a leading indicator for technology. Whatever people are doing for fun becomes something that companies later do for profit.

Most large service providers are feverishly working to optimize the performance of interactive games - which means understanding and implementing effective QoE network mechanisms. Vendors are developing monitoring and measurement tools that will let service providers offer QoE guarantees, which will make it possible for companies to roll out interactive applications above and beyond the entertainment variety.

The third development is storage-area networks and network-attached storage across the WAN. Companies today are experiencing two key business drivers: reducing costs and reducing risk. Many are consolidating facilities to cut costs, only to discover they've increased their reliability risks. Consolidating from 10 data centers to two increases the likelihood that a catastrophic network or server outage could take down the company.

That's why many folks are investigating remote storage facilities, which require high-bandwidth links and extremely high reliability. We know that the storage service provider (SSP) idea has been tried and failed, but we aren't talking about outsourcing the storage management. Companies are still plenty interested in networks capable of providing access to their own remote storage facilities - and that represents a potential bonanza for network providers.

Johnson is senior vice president and CTO for Greenwich Technology Partners, a network consulting and engineering firm. Her column appears biweekly. She can be reached atjohna@greenwichtech.com.

More Eye on the Carrier columns