New tools nix performance glitches

IT managers are too often the last to know that something has gone wrong. New performance management technologies, however, are changing that. For a closer look at tools that put managers in the know, searchWindowsManageability interviewed Diane Hagglund, director of product marketing for Sunnyvale, CA-based Mercury Interactive. Mercury's Topaz provides an user's-eye-view of application availability.

Like jilted lovers, IT managers are too often the last to know that something has gone wrong. The day is going fine, for example, and then the phone lines start buzzing with customers reporting a malfunctioning Web page. Fortunately, new technologies make it possible for IT administrators to be the first to know about system glitches and to fix them before customers are affected. For a closer look at technologies that put managers in...

the know, searchWindowsManageability interviewed Diane Hagglund, director of product marketing for Sunnyvale, CA-based Mercury Interactive. Mercury's Topaz 3.5 software, starting at $995 and released in July, provides an user's-eye-view of application availability.

sWM: How has application performance technology changed in the Internet age?
Hagglund:

The traditional way of making sure that mission critical applications are running has been to look at the individual components in those applications. That's monitoring the servers to make sure they're running and monitoring the network devices to make sure they're up and then making the assumption that if all these pieces are working, the end-user is OK. That was an assumption that worked really well in the days of the mainframe, because there were only a couple of things to monitor. With complex Web applications, that approach is no longer adequate. It's very important to monitor and manage these applications from what the end-user is experiencing, asking yourself: "Is my application up and down and performing well for the guy who's job it is to use this application?"

sWM: What action items would you put on an IT manager's daily performance management agenda?
Hagglund:

Look at your end-user. That's what matters more than anything else. Make sure the scripts you're writing represent your end-user. If 90% of your users are connecting using 13.4 modems, you have to understand that and emulate that. Definitely make sure your alerts are meaningful. Match your monitoring to your business needs.

sWM: How does Mercury's Topaz technology deliver end user-focused information?
Hagglund:

Topaz creates a script that represents what a real user would be doing, such as a transaction on the site. For example, that transaction could be transferring money from a checking account to a brokerage account. Emulating a real user includes all aspects of the customer experience, from the type of browser they're using to the execution of any client side technologies.

sWM: Could you offer an example of how an IT manager would use Topaz to gather performance metrics?
Hagglund:

On a live production system, you use a script to measure the experience the end-users are having. For example, as a first step, you can determine that the process of a customer support rep searching for a customer name, in order to call him back, is actually taking him a specific amount of time. Then, once you can measure that, the second step to set an objective for what acceptable performance is. You may say it is acceptable to have a four-second response time for a customer search and use those requirements to set a threshold. If it takes longer for the application to work, Topaz will send out an alert. The third step occurs when those thresholds are exceeded. Topaz can send out an alerted via a pager, e-mail or an SNMP trap to the management console. The fourth step, the most important piece, is fixing the problem once you know you have one. Topaz provides a suite of diagnostic tools that help you pinpoint the root cause of performance problems after an alert has been sent out.

sWM: How do Topaz' diagnostic tools pinpoint problems?
Hagglund:

Topaz' technology can take a transaction and break it down into where the transaction is being spent. You can look at a transaction that has taken 10 seconds and see right away that two of those seconds were in the network and eight of those were in the server. So you know to page the internal workers because you know right away that's where your problem is.

sWM: Without using a technology like Topaz, how could an IT manager check on application performance?
Hagglund:

There is the ability to ping the Web site. (Ping stands for Packet Internet or Inter-Network Groper, a basic Internet program that lets you verify that a particular IP address exists and can accept requests.) You can measure using a very basic technology like ping and find out that a Web page is up or down or is responding to ping. This approach doesn't measure customer experience very effectively, however. The page could be downloading in one second and saying "cannot log on." That's not a good customer experience.

Dig deeper on Enterprise Infrastructure Management

Pro+

Features

Enjoy the benefits of Pro+ membership, learn more and join.

0 comments

Oldest 

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to:

-ADS BY GOOGLE

SearchServerVirtualization

SearchCloudComputing

SearchExchange

SearchSQLServer

SearchWinIT

SearchEnterpriseDesktop

SearchVirtualDesktop

Close