Converged Application & Infrastructure Performance Monitoring

• Simulate user sessions to key business services and measure end-user experience
• Monitor availability and performance of websites/web applications 24x7
• Test with simulated user transactions to establish baselines of web performance
• Proactively get alerted to user experience issues before business impact
• Measure user experience even when no real users are accessing the application
• Benchmark user experience metrics and compare with different locations

Key performance questions answered:

• Are the critical business applications available to handle user requests?
• What is the total response time for a typical user access to a business-critical service?
• Which web pages/URLs are slow or failing?
• Are there specific times of day when the slowness occurs?
• How does the performance vary depending on the geographic location from which the user is connecting?

• Monitor user experience of end-users in real time as they access a website or web application
• Pinpoint which users from which are impacted by poor application performance
• Track user satisfaction using industry standard Apdex score
• Monitor page load times and proactively get alerted to slowdowns, JavaScript errors, and downtime
• Break down page load time into browser initial request, network time, server time, content download time and browser response time
• Identify slowdown due to DOM download, high DNS lookup time or TCP connection time
• View the breakdown of all page resources to gauge the impact of JavaScript, CSS, and images, as well as third-party content such as ads, social, and analytics
• Monitor real user experience of base pages, iFrames and Ajax pages
• Analyze traffic to web pages and identify traffic peaks
• Slice and dice historical data to view user experience metrics by end-user device, by browser, by website, by page type, by page group, by city, by region, and so on

Key performance questions answered:

• What are the top 5 poorly performing page groups?
• Where is most of the traffic coming from—which city, country, browser, device type (mobile/desktop/tablet)?
• Geo performance: Are certain locations slower versus others?
• Are my users satisfied, tolerating or frustrated?
• Did a site slowdown coincide with a peak in traffic?
• Which page types are most impacted and performing poorly: base page, Ajax, iframe?
• Are users seeing any JavaScript errors and on which pages?
• When a page is slow is the issue due to the client, or server, or network, or content?

• Perform distributed transaction tracing for application transactions all the way through the application framework to the database and back
• Perform tracing without requiring any changes to the application code
• Tag-and-follow tracing enables seamless tracking of transactions as they pass through multi-tiered application stacks
• Support Java and Microsoft .NET based applications
• View on intuitive visual topology map the application runtime path for each request and the processing time spent at each tier
• Isolate problematic Java/.NET code that is causing errors or exceptions
• Identify methods that are responsible for high request processing time
• Drill down to the specific database query that is responsible for database slowness
• Track down specific HTTP, JMS, WCF or web services calls that are slow
• Quantify enterprise application responsiveness by tracking calls to SAP JCO, Jolt and other application APIs
• For .NET applications, monitor if ADO.NET or ODP.NET requests to the database are slow
• Monitor transactions from both web browsers and native mobile applications

Key performance questions answered:

• What URLs are being accessed and what are their response times?
• Are any transactions having errors?
• Which transaction is slow to respond and why?
• Is slowness due to a code-level issue, query-level issue or slow remote call execution?
• Which part of the application code is taking high processing time?
• Which query is executing slowly?

• Track JVM/CLR CPU utilization and easily identify high CPU consuming threads
• Monitor heap and non-heap memory usage (growing memory utilization can indicate memory leak and out-of-memory errors)
• Monitor JVM/CLR uptime statistics and whether restarts are occurring unexpectedly
• Identify class names that are memory leak suspects
• Uncover deadlocked and blocked threads, and easily isolate the Java class, method or object that is causing these issues
• Look up the stack trace to pinpoint the exact line of code encountering an error or exception. See real-time as well as historical data.
• Identify times when garbage collection is taking too long and adversely affecting Java application performance

Key performance questions answered:

• Is there any runaway thread hogging the CPU?
• Which line of code is it executing, in which class, and which method?
• Is the JVM/CLR heap and non-heap memory sized correctly?
• Are there any out-of-memory-exceptions or memory leaks?
• When does garbage collection happen, and how much memory is freed up each time?
• Are there any thread deadlocks causing application processing to hang?

• Monitor the number of incoming requests by each connector
• Monitor the connections available in the DataSource connection pool
• Track how long each servlet takes for execution
• Measure invocation, execution, creation, removal metrics for each EJB
• Identify if the EJB thread pool is sized correctly
• Diagnose any waiting threads in JCA Connection Pools
• Track JDBC connectivity and the requests waiting to be processed

Key performance questions answered include:

• Is the EJB thread pool sized correctly?
• Are there too many requests coming into the web server?
• Are any connections being dropped?
• Are there any connection leaks in JDBC connectivity?
• Are there enough threads to process incoming requests?
• Are Java servlets executing within acceptable processing thresholds?
• Are messages waiting too long in the queue to get processed?
• Is a backlog accumulating on a particular JMS queue or topic?
• Are transactions waiting for connections from the connection pool? Is it sized optimally?

• Server
• Operating system
• Virtual machines and hypervisors
• Containers and orchestration platforms
• Databases
• Web servers
• Messaging servers
• Enterprise Service Bus
• Active Directory/ DNS/DHCP
• Cloud infrastructure
• Network devices (routers, switches, firewalls, load balancers)
• Storage systems
• Virtual desktop infrastructure

eG Enterprise enables IT teams to monitor all these IT infrastructure components in context of application performance. Get converged visibility of apps and infra from a single pane of glass. Key performance questions answered include:

• Is a physical/virtual server running out of resources?
• Are there any network delays, packet drops or bandwidth congestions?
• Is there any authentication issue in Active Directory?
• Is the database incorrectly sized or designed?
• Are there any connection leaks in JDBC connectivity?
• Are there any storage IO hotspots?
• Is there a problem on container creation due to resource constraints?
• Are cloud services healthy and working fine? Is there any cloud outage affecting application performance?

• Auto-discover and visualize application dependencies with underlying infrastructure
• Auto-construct a service topology with the application and supporting infrastructure
• Auto-correlate performance alerts across applications and infrastructure tiers for root cause diagnosis; enhance service uptime and performance and reduce mean time to repair
• Automatically baseline performance metrics using machine learning
• Automatically track configuration changes to applications, operating system, devices to see if an config change is impacting application performance
• Leverage out-of-the-box and custom reports (historical analysis, comparison, optimization, right-sizing)