ApplicationVantage

Pinpoint application performance problems

ApplicationVantage is the premier application performance troubleshooting solution in the marketplace today for addressing application transaction performance concerns. It helps find and fix application performance and infrastructure problems in production and pre-production environments. By providing detailed application troubleshooting insight, ApplicationVantage pinpoints and corrects the causes of poor end-user response times wherever they hide, which eliminates time-consuming guesswork. With its unique predictive analysis features, you can see the impact of various performance adjustments on a transaction's response time before taking action. ApplicationVantage helps ensure new and modified applications roll out successfully the first time around, and provides crucial information for establishing and meeting application service requirements.

With ApplicationVantage, you can:

• effectively troubleshoot application performance problems in pre-production and production environments with a single product
• isolate application performance problems to client workstations, the network, a server, or the application itself
• validate corrective actions using unique predictive analysis features
• certify the success of new and modified application deployments
• help establish and meet service requirements.

The Vantage Packet Trace Analysis Report helps identify network problems quickly by breaking out the lower layers of network traffic.

The optional Unattended Capture Manager (UCM) module allows administrators to manage the capture of network traffic at strategic points in the network and retrieve the captured data on an "as needed" basis. When collecting the data from remote agents, a time filter for a specific period can be selected to limit the amount of data captured. For example, agents can be set up to capture data for three hours. If a problem occurs during that period, the administrator may decide to retrieve data five minutes before and after the event so only the relevant data necessary to troubleshoot the problem is collected.

The Extended Merge allows the user to merge any number of trace files together to provide a unified view of the transaction. The Extended Merge module provides remote capture capabilities for specific problem transactions. Users can set up filters to obtain only the relevant data and perform the start and stop of the capture in a controlled environment. From there, they can retrieve and merge any number of traces to obtain a realistic view of the transaction and determine the source of the problem.

The Expert Analysis makes it easy for the user to solve problems quickly and accurately. This module uses a proven troubleshooting methodology to guide users through the problem resolution process so that problems are addressed quickly and accurately, often using less experienced personnel.

The combined Conversation Map and Bounce diagram allows IT professionals to pinpoint what packets the network is delaying.

Understand a distributed application's behavior and performance dependencies

The application Thread Analysis, for example, provides an easy way to understand the files or commands an application sends over the network in a single transaction, such as Java commands, SQL statements or HTTP commands. It also provides insight into the time and duration of a thread and the relationships between threads. Optional decode modules analyze specific database and middleware commands to isolate performance problems.

Thread Analysis supports:

• Databases: Microsoft SQL Server, Oracle, Sybase, Informix and IBM UDB (Version 8 and above). All databases require separate modules.
• Middleware: IBM WebSphere MQ
• Web-based applications: Web services using Java (RMI/IIOP) and Java scripts, SOAP, XML, HTTP, HTTPS, HTML, NNTP, JPEG/GIF files, SMTP and POP3.
• Others: DCOM, SSL, NTLM, Microsoft Networking (SMB), Citrix (packet-level decodes), Generic SQL, Novell NCP, NFSv2 and RPC.

The Performance Overview provides a summary of how the transaction is behaving. It helps determine if a performance problem is due to the client, the network, a server, or the application itself.

The versatile CNS breakdown allows the user to quickly identify the most significant contributor to poor performance.

Predictive Analysis Troubleshooting means not only identifying the cause of a problem, but also a recommendation on how to fix it.

Vantage provides a predictive analysis feature that allows users to see the impact of various performance adjustments on a transaction's response time. In this way, the user can validate that performance will be improved, and by how much.

Users can see the effects of varying the following metrics: Network: modifying bandwidth, latency and load Server: increasing/decreasing the power of a server Application: changing application turns for individual "thread" components, varying TCP window size By adjusting these parameters one at a time, users can determine if the overall response time of the transaction is affected significantly. If it is, they can recommend a corrective course of action that will resolve the problem.

Vantage's Predictive Analysis let's you see the effects of changing network, node, and application parameters to help validate corrective action to performance problems. This example shows the effects of changing bandwidth.

The optional WAN Deployment Expert (WDE) module focuses on the dominant network budget metric: WAN bandwidth. It allows users to determine what impact hundreds of users will have on their infrastructure before rolling out a new application. With WDE, users can use the previously captured tasks in Vantage to quickly build typical workload scenarios of users at remote branch offices. From there, they can help ensure application deployment success.