|
SYS-CON.TV Webcasts
Comments
Did you read today's front page stories & breaking news?
SYS-CON.TV
|
Top Links You Must Click On
Product Reviews VERITAS I3 for J2EE
VERITAS I3 for J2EE
By: Rob Halleron
Aug. 5, 2004 12:00 AM
Sometimes as J2EE application developers we feel like we are in a darkened room. We know that something is wrong with our application, but we have no idea where the problem is. Application performance management (APM) tools, such as VERITAS i3 for J2EE, has helped us "turn on the lights" by enabling us to see exactly where in the application our problem really is. Once we identified the problem, it all flowed from there, as we could look at how the problem affected our application from end to end and make the right decisions on how to fix the problem. VERITAS i3 APM software is the only solution we found that provides such end-to-end application visibility. VERITAS i3 for J2EE VERITAS i3 for J2EE can quickly, efficiently, and unobtrusively capture the metrics necessary to appropriately tune J2EE-based applications. It presents these important metrics in a manner that enables crisp communication, rapid detection, correction, and verification throughout the application's life cycle. Installing and Using VERITAS i3 VERITAS i3 for J2EE provides great visibility into application performance problems through a GUI that lets you drill down from an alert to where the problem lies. For example, it understands response time contributions from Java servlets, JSP, EJBs, JMS, JNDI, JDBC, and XML. It correlates activity across Web, multiple JVMs, and DB servers. It also has a SmarTune feature that gives you great advice on how to fix the problem. For example, a third party wrote part of our application that served up static content about cruise holidays. These pages should have been delivered fast, since they can be stored in cache memory. Using VERITAS i3 we found the problem was that the application was making a database call for each statement asking for content. The product allowed us to find and fix that problem quickly. In another instance, we were able to identify poorly performing SQL statements, including one particular query that was running at 0.5 of a second but was occupying one entire processor. We were able to tune this query down to 0.08 of a second. Starting at the Insight screen in Figure 1, there is an overview of the three layers to the system: Web, application, and database servers. From the graph on the left, most of the time is spent in the J2EE layer. We could investigate that further by choosing the J2EE option on the top menu. In Figure 2, the method invocation graph appears to show two high usage items but these are part of struts and so will normally be high. However, the third item is the logon process and should be quick, so this could be investigated further. The JVM etailJVM6 is also more heavily loaded than the others, which may indicate a balance problem. Clicking on the third item in the method invocations graphs gives more details (see Figure 3). Clicking on the top item in the list digs into that particular call to reveal these sub calls (see Figure 4). Most of the time is spent local to the routine com.tuiuk.etail.channel.shop.agentlogon.servlet.AgentLogonServlet.service. A developer can now investigate why it is using up the majority of the response time. If we take a step back to Figure 1, we can investigate the top Oracle statement in the graph. Clicking on the top item in the bar graph and then the Oracle tab in the top menu bar takes us to the screen in Figure 5. We can now launch Indepth for Oracle to determine what the statement is. As you can see in Figure 6, this is a very large INSERT statement that is part of our content-refresh process, so it's not unreasonable for it to take a while to process; nothing to worry about there. Summary SIDEBAR JDJ Product Snapshot Target Audience: Java application architects/developers and application managers Con: SIDEBAR 2 VERITAS Software Corporation 350 Ellis Street Specifications Operating Systems: Sun Solaris 2.6, 7, 8, 9; IBM AIX 4.3.3, 5.1, 5.2; HP-UX 11.0,11i; Windows NT SP6a, 2000 SP3; Linux Red Hat 7.2, 8 Advanced Server 2.1; SuSE Linux 8.0, Linux S/390 Pricing: Based on number of processors and server class. Test Environment Reader Feedback: Page 1 of 1
Enterprise Open Source Magazine Latest Stories . . .
Subscribe to the World's Most Powerful Newsletters
Subscribe to Our Rss Feeds & Get Your SYS-CON News Live!
|
SYS-CON Featured Whitepapers
Most Read This Week |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||