Article

Data compression puts auto parts maker in fast lane

Jim Rendon, News Writer

When the network link between O'Reilly Automotive Inc. and its data center reached its capacity limit, the company didn't buy more bandwidth, as most businesses would. Instead, O'Reilly compressed its data stream.

O'Reilly, an auto parts dealer based in Springfield, Mo., came across a low-cost network service that companies are increasingly turning to: bandwidth optimization. O'Reilly has about 1,100 retail stores, 10 distribution centers and a disaster recovery data center, which is located in Texas.

The company had been using a 12 Mbps link to transport data to the disaster recovery center, but last spring the company began to experience problems with latency. Between point of sales applications, inventory management systems, e-mail and legacy applications, the company was generating 350 million transactions a week, said Mark Garton, disaster recovery team leader at O'Reilly.

The company wanted to ensure that it would lose as little data as possible if the system ever went down, Garton said. A business partner sent O'Reilly to see network optimization vendor Expand Networks Inc.

Expand, based in Roseland, N.J., produces a series of appliances that use standards-based compression to speed the movement of data across networks.

Pedro Colaco, vice president of marketing at Expand, said that the appliances give companies between two and five times more capacity on a wide area network link. With such a drop in congestion, latency improves

    Requires Free Membership to View

    Login

as well.

Last spring, O'Reilly installed Expand's Accelerator 6810 series appliance. Since then, Garton said, the company hasn't had any problems with link capacity or latency. In fact, it has been able to switch from a nearly full 12 Mbps link to a 6 Mbps link that is only at 50% capacity.

Plus, the company is saving $32,000 annually on bandwidth costs. O'Reilly is currently testing the product for further use on its main network.

Expand leads the bandwidth optimization market. It was the first company in that space in 1998, said Peter Firstbrook, a senior analyst with Stamford, Conn.-based Meta Group. But the market is still young, and a number of other vendors are now gaining a foothold, including Peribit Networks Inc., Packeteer Inc. and FatPipe Networks Inc.

Each of these companies uses a slightly different approach to optimization, Firstbrook said. For instance, Expand uses standards-based algorithms at layer 7, while Peribit uses an algorithm that was designed to identify patterns in gene sequences.

Dialog: Data model discussion

Chris: One thing that has us all puzzled is exactly how the ITRP concepts fit together and work with other non-ITSM concepts. There's a lot of terminology, and it seems like things overlap sometimes. For example, what is the relationship between a Configuration Item and an Asset? Also, some of what ITIL calls for is not exactly how we do business. Do all Configuration Items go through our data center change control process? What is the relationship between a Service Request and an Incident? Is a Service a Configuration Item? Is a Service Offering? Are Applications Services?

Kelly: That's why we're going to turn to one of the most important aspects of enterprise architecture: the creation of a conceptual data model.

Chris: A conceptual data model? What good is that? We're probably not going to build anything—we're going to purchase products. Sounds pretty technical.

Kelly: That's why I call it a conceptual data model, and yes, it's relevant even if you are purchasing products. There are a lot of vendors out there selling various fl avors of IT enablement and IT governance tools, and they have a lot of overlap between their products, often with slightly different terminology.

A conceptual data model is not technical—it's about clarifying the language describing our problem domain so that we understand exactly what we mean by a Configuration Item and how it might relate to a Service. And this is something you need to put together independent of the products— because it's going to be your road map that helps you determine what products you need.

Chris: Will it help me translate the vendor-speak?

Kelly: Absolutely. One vendor may have a "service catalog entry" and an "order," and another vendor may call the same two things a "template" and a "service instance" In the conceptual data model (also called a "reference model"), they are Service Offering and Service. It doesn't matter what the vendors call them, but you need to understand that any service request management solution should have both concepts. Doing the data model helps us understand our requirements better and communicate them to the vendor.
It's hard to tell which approach will work best for a company's network, Firstbrook said. Often, the answer depends on the kind of traffic on the network. To find the best approach, he said, a company should test the top vendors' products on its network to see which works best.

But these products are not for everyone. Expand charges uses based on throughput, and it costs $2,000 to compress a 128 Kpbs link. Compression of a T3 line costs $45,000.

Firstbrook said that these services often appeal most to those companies with significant bandwidth needs, as well as enterprises that do business overseas, where bandwidth provisioning can often be expensive and time-consuming.


Join the conversationComment

Share
Comments

    Results

    Contribute to the conversation

    All fields are required. Comments will appear at the bottom of the article.
    Back to top