Reference Graph Tree | HP jmeter Software for -UX instruction

Unfinalized Objects

The number of objects with pending finalization, arranged by their type and sorted by the number of objects of each type. These are all objects with the finalize() method available, but not yet invoked.

Related Topics

•Mark an Item for Search (page 178)

•Find a Search Pattern (page 179)

•Threads Histogram (page 152)

•Unfinalized Objects (page 126)

•Unfinalized Queue Growth (page 115)

Reference Graph Tree

Shows all live objects and references between them.

Expanding the nodes shows which references a given object holds.

HPjmeter uses heuristics to detect soft or weak references and uses special symbols ~->for them, while the hard (regular) references are marked by ->. Double-click on an object to view additional information: all references to the object, and the number of bytes held in the heap by this object. This number is the total amount of bytes that could be reclaimed by garbage collection if all references to this object were nullified. This piece of information is sticky — once this value is calculated for an object, it will be displayed in the Reference Graph.

Displaying large reference graphs puts a lot of stress on the underlying Swing components, so please be patient when dealing with huge heap dumps.

Related Topics

•Mark an Item for Search (page 178)

•Find a Search Pattern (page 179)

•Class Loaders (page 150)

Reference Sub-Trees by Size

This visualizer shows the cluster of objects that are directly or indirectly referenced from the root of the tree, along with the cumulative size of all the references.

This metric is similar to Reference Graph Tree, with the following differences:

•You can see the amount of memory allocated to a sub-tree(a node and its contents) as well as to the node itself. The memory held by the node appears in parentheses ( ), and the memory held by the sub-tree appears in brackets [ ].

•The table on the right side displays object and primitive values for any selected node object in the tree.

In Figure 8-19, an object of type VecTest3 is selected in the tree. The table on the right is showing the Attributes tab, which displays a list of the non-static fields in the object.

146 Using Visualizer Functions

Image 146

HP jmeter Software for -UX manual Reference Graph Tree, Reference Sub-Trees by Size

Contents

HPjmeter 4.2 Users Guide Document Notice Table of Contents Monitoring Applications Analyzing Garbage Collection Data 116 Unfinalized Objects Live Objects Bytes 197 187192 199 Typographic Conventions About This DocumentIntended Audience Additional HPjmeter Documents Related Information Publishing History HP Encourages Your Comments Features Introducing HPjmeter Node Agent ConceptsJVM Agent Agent Requirements Completing Installation of HPjmeterPlatform Support and System Requirements Console Requirements Attaching to the JVM Agent of a Running Application Completing the installationFile Locations Preparing to run Java Example Usage On Java Selecting Other JVM Agent Options JVM Agent OptionsShowing Version Information Include=filter1filter2..., exclude=filter1filter2 To specify a file name Noalloc Working with Firewalls Security AwarenessJVM Options Usage Examples Configuring User Access Using HPjmeter to Monitor Applications Getting Started To connect to a node agent via a SecureShell SSH tunnel Using HPjmeter to Monitor Applications Set Session Preferences When the Optional Port Box Is Not Used Getting Started View Monitoring Metrics During Your Open Session Using HPjmeter to Analyze Profiling DataChanging Session Preferences During a Session Configure your application Using HPjmeter to Analyze Garbage Collection Data Monitoring Demonstration Instructions Memory Leak Applications Thread Deadlock Sample See also Heap Usage Notification Thread Deadlock Example Display Controlling Data Collection and Display Setting Data Collection PreferencesMonitoring Applications Running Node Agent as a Daemon Managing Node AgentsManaging Node Agents On HP-UX Verifying HP-UX Daemon is Running Running Multiple Node Agents Node Agent Access RestrictionsSaving Monitoring Metrics Information Stopping Node Agents Saving Data from the Console Diagnosing Errors When Monitoring Running ApplicationsNaming Monitoring Data Files Identifying Unexpected CPU Usage by Method Checking for Long Garbage Collection Pauses Checking for Application Paging ProblemsViewing the Application Load Identifying Excessive Calls to System.gc Reviewing the Percentage of Time Spent in Garbage Collection Monitoring Applications Determining the Severity of a Memory Leak Checking for Proper Heap SizingConfirming Java Memory Leaks Identifying Abnormal Thread Termination Identifying Excessive Object AllocationIdentifying the Site of Excessive Object Allocation Identifying Multiple Short-lived Threads Identifying Excessive Thread Creation Identifying Excessive Lock ContentionIdentifying Deadlocked Threads Identifying Excessive Method Compilation Example Metric Method Compilation Count Using the JMX Viewer Identifying Too Many Classes Loaded Understanding the JMX Summary View JMX Summary Tab Appearance of the Summary JMX Viewer When First Opened JMX Memory Tab JMX Viewer with Summary Memory Tab Selected JMX Threads Tab JMX Viewer with Summary Threads Tab Open JMX Notifications Tab Changing Mbean Values and Monitoring the ResultJMX Runtime Tab Using the Functions in the JMX Server View MBean Filter MBean Attribute Tab MBean Operations Tab 10 MBean Operations Tab Open for Display MBean Notifications Tab MBean Information Tab 12 MBean Information Tab Open for Display Profiling Applications Profiling Overview Sampling Tuning PerformanceTracing Preparing a Benchmark Collecting Profile Data Profiling with -Xeprof If timeon=sig... has not been specified, the default is Profiling with Zero Preparation Profiling with -agentlibhprof Following options are useful Supported -agentlibhprofoptions Naming Profile Data Files Generate more complete call graphs Inclusive Method CPU Times Profiling with -Xeprof Profiling with -agentlibhprof Looking at the Data from the Bottom Up Approaches to Analyzing Performance DataConsiderations in Interpreting the Data Looking at the Data from the Top Down Time Units Locating Summary Information for Saved Data SetsInclusive Versus Exclusive Time CPU Versus Clock Time Adjusting Scope Estimate menus Comparing Profiling Data Files Scaling Comparison Data Reading Profiling Histograms Key to Thread States Reported by Interpreting the Histogram Presentation Using Call Graph Trees Interpreting Call Graph Data Example of Node Color Display Tree Pruning Using Sub-TreesOptions for Manipulating the Call Tree Display Auto-Expanding the Call Tree Using Heuristics to Locate Possible Hot Spots Searching the Trees Using Heuristics to Locate Possible Hot Spots Data Collection with -Xverbosegc Analyzing Garbage Collection DataObtaining Garbage Collection Data Java ... -Xverbosegchelp 13, Old generation, in bytes. These spaces are reported Collecting GC Data with Zero Preparation Collecting Glance Data for Viewing in HPjmeter Data Collection with -Xloggc Related Topics Using -XX+PrintGCDetails and -XX+PrintHeapAtGC Together Naming GC Data Files Comparing Garbage Collection Data Files Basic Garbage Collection Concepts Key to Garbage Collection Types Recognized by HPjmeter Understanding the Summary Presentation of GC Data Common Garbage Collection Types Recognized by HPjmeter Summary Panel Showing Garbage Collection Statistics Understanding the System Details Captured with GC Data Basic Garbage Collection Concepts Analyzing Garbage Collection Data Starting the Console Using the ConsoleUsing the Main Window Functions Data Representation Main Console Window JVM Agent Icons and Their MeaningNode Agent Saving Data Open and Cached SessionsTime Slice Entries Console Tool Bar Buttons Console Menu Choices Monitor Menu Remember Main Window Location Console Guide Console Guide Location and First Screen Specifying Metrics to Collect for Monitoring Setting Monitoring Session PreferencesStatus Bar Enabling Monitoring Alerts Adding Filters Specifying Filters for MonitoringEditing Filters Specifying Filter Sets Using AlertsViewing Monitoring Data in HPjmeter Appearance of Alert Threshold in GC Duration Visualizer Using the Alert Controller Appearance of Alert Notices in the Main Console Pane Alert Controller Window Editing E-mail Notification Attributes Viewing a Log of the Alert History Responding to Alerts Abnormal Thread Termination Alert Identifying Abnormal Thread Termination Thread Histogram GC Duration Notification Expected Out Of Memory Error AlertExcessive Compilation Alert Heap Usage Notification Java Collection Leak Locations Alert Array Leak Locations Alert 13 Java Collection Leak Locations Visualizer Process CPU Usage Alert 14 Array Leak Locations Visualizer System CPU Usage Alert Thread Deadlock Alert Unfinalized Queue Growth Using Visualizer Functions Profiling Data Viewer Using Monitoring Displays Locating Information About a JVM and its Environment Monitor Code and/or CPU Activity Menu Java Method HotSpots Details Thrown Exceptions Thrown Exceptions with Stack Traces Monitoring Metric Thrown Exceptions with Stack Traces Monitor Memory and/or Heap Activity Menu Heap Monitor Garbage Collections Monitoring Metric Garbage Collections GC Duration Percentage of Time Spent in Garbage Collection Unfinalized Objects Allocated Object Statistics by Class Allocating Method Statistics Current Live Heap Objects Name of class to which object belongs Monitor Threads and/or Locks Menu Thread Histogram Already acquired by another thread Lock Contention 14 Monitoring Metric Lock Contention Monitor JVM and/or System Activity Menu Method Compilation Count Method Compilation Frequency Loaded Classes Percent % CPU Utilization 17 Monitoring Metric Loaded Classes Related Topics Using Profile Displays Viewing Profiling or GC Data in HPjmeter Using Visualizer Functions Inline Candidates Exceptions Thrown Memory Leaks Method Call Count Menu ChoicesProfile Code and/or CPU Activity Call Graph Tree with Call Count Exclusive Method Times CPUExclusive Method Clock Times Inclusive Method CPU Times Call Graph Tree with CPUCall Graph Tree with Clock Time Inclusive Method Clock Times Average Inclusive Method Clock Times Average Exclusive Method Clock TimesAverage Inclusive Method CPU Times Starvation by Method Methods with Loops Profile Memory and/or Heap ActivityStarvation Ratio Exclusive Class CPU Times Created Objects Bytes Objects Created by MethodCreated Objects Count Live Objects Count Reference Graph Tree Reference Sub-Trees by Size Using Profile Displays For string and char arrays, see String and char arrays 21 Long String Pop-up Box Class Loaders 23 Class Loaders Visualizer 24 Class Types Pop-up Window Guidelines Residual Objects Bytes Profile by ThreadsResidual Objects Count Threads Histogram Profile by Locks Thread Groups Histogram Contested Lock Claims by Method Lock Delay Method ExclusiveLock Delay Call Graph Tree All Lock Claims by Method Average Exclusive Method Lock Delay Lock Delay Method InclusiveLock Contention Ratio by Method Exclusive Method Lock Delay / Clock Time Inline Candidates Using Heuristic Metrics from the Estimate MenuExclusive Class Lock Delay Exceptions Thrown Using Specialized Garbage Collection Displays Heap Usage After GC Duration Stop the World 27 GC Metric Heap Usage After GC Related Topics Cumulative Allocation 28 GC Metric Duration Related Topics Creation Rate 29 GC Metric Cumulative Allocation Related Topics Allocation Site Statistics 30 GC Metric Creation Rate Related Topics 31 GC Metric Combined Sets of Allocation Site Statistics 32 GC Metric Separated Sets of Allocation Site Statistics Using Visualizer Functions Filter Field Descriptions Using Visualizer Functions User-Defined X-Y Axes Multiple User-Defined 37 User-Defined Comparisons of GC Metrics Related Topics 38 Multiple User-Defined Metrics Glance Data 40 GC Metric Glance Data 41 Glance Data Metric Configuration Box Run Queue Average Glance System Call Data Are private 42 GC Metric Glance System Call Data Using Visualizer Tool Bars Common Tool Bar Buttons Tool Bar Buttons for Manipulating Garbage Collection Data Tool Bar Buttons for Manipulating Tabular DataTool Bar Buttons for Manipulating Graphical Data Special Button or Other Gadget Functions Mark an Item for Search Find a Search Pattern 44 Locating an Item Pause or Resume Graphical Time-based Scrolling Changing Time Interval in GC Data Visualizers 46 The View Menu Selections in the GC Viewer Using Visualizer Tool Bars Changing Time Interval in Monitoring Visualizers Change Color Selection for Histogram Display 50 Time Interval Use in a Monitoring Visualizer Thread Histogram Profile by Threads Understanding How HPjmeter Works Performance Overhead on Running Applications Console Overhead Using Confidence Interval to Indicate Sample ValidityData Sampling Considerations How Memory Leak Detection Works Related Topics Page Documentation and Support TroubleshootingInstallation Identifying Version Numbers Console See also To Take Advantage of Dynamic Attach If you see this JVM agent error on startup JVM agent Node agent Zero Preparation Profiling Unexpected Behavior in Visualizers Connecting to the HPjmeter Node Agent Quick References Start and stop verbose GC data 0.14 0.02 Active mode Glossary Jvmpi SSL Symbols Index Alert Notification Editor, 107 alerts Restore Roots Alert Notification Editor, 108 count Collecting with zero preparation, 83 GC event Glance system call data, 174 graphical scrolling JVM Duration Stop the World, 158 enable, 99 User-Defined X-Y Axes for GC analysis, 167 viewing, 29 Program termination, 47 overhead Scale Special, 72 scaling data Xeprofoptions Xbootclasspath