Taking Steps

By David F. Carr  |  Posted 2006-12-18 Email Print this article Print
 
 
 
 
 
 
 

An overloaded network infrastructure was making employees at Inergy Automotive Systems cranky. Upgrading the operating system and prioritizing data transmissions increased bandwidth by 300%—-and quelled complaints.

Taking Steps

In the first phase of Inergy's network integration project, which lasted from June to December 2004, DeSouza and his team eliminated redundant network servers and services. One of the advantages of the Active Directory architecture is scalability. Whereas the Windows NT generation of Microsoft's network operating system essentially requires each location to be treated as a separate network domain, Active Directory made it possible to operate the entire network as a single domain with a smaller number of servers acting as domain controllers. In Microsoft's network architecture, a domain controller is the server that controls log-in requests and network access permissions. Most Inergy branch locations now have a single domain controller, and that server often doubles as a file and print server if it's not too heavily loaded.

Instead of installing a Microsoft Systems Management Server (SMS) at each location, Inergy moved to a central installation of SMS and the Microsoft Operations Manager for network control. Inergy also upgraded to a single backup system, which replaced the mishmash of backup systems that had previously been deployed, and it outsourced its e-mail servers. For its backup software, Inergy chose Veritas, which was acquired by Symantec last year.

All told, Inergy was able to cut the number of servers on the network from 220 to 160, DeSouza says. Having fewer servers also lowered the manpower required for network administration, particularly after Inergy added Quest Software's ActiveRoles Server for simplified Active Directory account creation and management in 2003. In all, the number of full-time equivalent employees required for domain administration dropped from 15 to three, he says.

Phase two of the network revamp began with a pilot project in early 2005; it used seven QOS appliances that DeSouza convinced Packeteer to lend him to demonstrate the value of the technology in practice. David Willis, a Gartner analyst, recommends this approach, saying most WAN acceleration vendors will lend equipment for a proof-of-concept project, which also helps enterprises measure the load on their networks.

QOS classification and enforcement capabilities have become commoditized recently as they've been absorbed into Internet routers from companies like Cisco. But Willis says there's still room in the market for vendors like Packeteer that have augmented QOS with data compression and protocol acceleration techniques to make WANs operate more efficiently.

"We still see about an 18-month gap in the functionality available through these WAN acceleration appliances and what you get with a router," Willis explains. Where Cisco has made inroads is by simplifying the process of classifying network traffic with an AutoQOS feature that, for example, is useful for automatically determining the right network settings for voice-over-Internet Protocol traffic, he says.

One obstacle to implementing QOS has been the complex and potentially political task of deciding which applications from which departments get the highest priority, Willis says.

According to DeSouza, Packeteer offered the more sophisticated level of QOS technology he wanted. For example, instead of assigning one flat priority to the Web's HyperText Transport Protocol, he wanted to give HTTP requests to the Inergy corporate portal a priority four times higher than Web access.

So far, Inergy has installed 36 of the Packeteer PacketShaper devices at locations in North America, Europe and Asia.

The QOS-enhanced network went live in the U.S. and Asia in the third quarter of 2005, and in Europe in early 2006. While it might sound like this was "just a box" to be plugged into the network, DeSouza says, "Nothing is as easy as it sounds." He traveled 22 weeks out of the year, working to keep the project on track, and some former colleagues burned out and left the company because of the aggressive schedule, he says.

Still, DeSouza is proud of the way the QOS project exceeded its objectives, coming in 30% under budget at a cost of 250,000 euros (about $315,000) as opposed to the estimate of 360,000 euros (about $450,000). At that rate, it paid for itself in about 18 months through tangible benefits such as avoiding the expense of buying additional WAN bandwidth from Inergy's telecommunications carriers. If intangibles such as the estimated benefit of improved employee productivity are included, the payback is only six months, DeSouza says.

Meanwhile, the effective bandwidth available on the network increased by about 300%, compared to the 200% improvement projected in DeSouza's original business case. The average transmission delay dropped by 75% (compared to a projected 50% improvement), and Web applications performed 50% faster (the projection was 33%). Bad transactions, such as failed postings to SAP, dropped by 20%; the goal was 10%.

DeSouza also sees another potential benefit for network security, since Packeteer bandwidth controls can be used to choke off network access to Internet viruses and worms that work by flooding a network with spurious requests.

Taken as a whole, the entire network integration project cost about $2 million, coming in about 15% under budget, with a 2.25-year payback based solely on the tangible savings.

The result is much different than what Inergy started with, taking a network from one corporate parent and a network from the other and mashing them together. Compared with that "administrative nightmare," says information systems VP Stephens, "Today, we have a first-class network that we're able to operate at a very low cost."

Even now, though he's happy when his CEO notices that there have been fewer complaints, Stephens is reluctant to boast. "When things go wrong, they tend to get noticed," he says. "When you start tooting your own horn, and then something does go wrong, that magnifies the issue." But for the moment, he can be happy that some things are going right.



<123>
 
 
 
 
David F. Carr David F. Carr is the Technology Editor for Baseline Magazine, a Ziff Davis publication focused on information technology and its management, with an emphasis on measurable, bottom-line results. He wrote two of Baseline's cover stories focused on the role of technology in disaster recovery, one focused on the response to the tsunami in Indonesia and another on the City of New Orleans after Hurricane Katrina.David has been the author or co-author of many Baseline Case Dissections on corporate technology successes and failures (such as the role of Kmart's inept supply chain implementation in its decline versus Wal-Mart or the successful use of technology to create new market opportunities for office furniture maker Herman Miller). He has also written about the FAA's halting attempts to modernize air traffic control, and in 2003 he traveled to Sierra Leone and Liberia to report on the role of technology in United Nations peacekeeping.David joined Baseline prior to the launch of the magazine in 2001 and helped define popular elements of the magazine such as Gotcha!, which offers cautionary tales about technology pitfalls and how to avoid them.
 
 
 
 
 
 

Submit a Comment

Loading Comments...

Manage your Newsletters: Login   Register My Newsletters