Borland Optimizeit Enterprise Suite 6

Transcription

1 Borland Optimizeit Enterprise Suite 6 Feature Matrix The table below shows which Optimizeit product components are available in Borland Optimizeit Enterprise Suite and which are available in Borland Optimizeit Suite: Optimizeit Optimizeit Optimizeit Thread Optimizeit Code Request Analyzer Profiler Debugger Coverage Optimizeit Enterprise Suite Optimizeit Suite Flexibility to support Java and J2EE development environments Improve the performance and reliability of any Java and J2EE code: Java applications, J2EE applications, servlets, applets, JavaBeans, Enterprise JavaBeans (EJB ), JSP applications, and Java tag libraries Multiplatform support for Windows, Solaris, and Linux. The Borland Optimizeit Suite 5.0 is also available for IBM AIX and HP-UX Integrates well with popular application servers including BEA WebLogic Server 6, 7, and 8; IBM WebSphere 5; Oracle9i Application Server; Borland Enterprise Server 5.2.1; JBoss 3.2; JRun 3.1; Sun ONE (iplanet) 6.5; and TomCat 3.3 and 4.1 Requires no code modifications; runs with most Java 2-compliant Virtual Machines (JVM ), including Sun J2SE v1.3 with builtin Java HotSpot Virtual Machine, Sun J2SE v1.4.1, and IBM Developer Kit, Java 2, Technology Edition, version 1.3 and supports BEA WebLogic JRockit 8.1 on Windows and Linux Support for leading IDEs, including seamless integration with Borland JBuilder, which enables developers to run Optimizeit Enterprise Suite right from within the JBuilder UI. Straightforward menu-driven integration with Borland Together ControlCenter, Eclipse, VisualCafe and IBM VisualAge. Integration is also available for IBM WebSphere Studio Application Developer (WSAD) NEW! Integrates with Eclipse 2, 2.1, and 3 Easily connects to a remote Java process to test a program running on a different machine 1

2 Borland Optimizeit Enterprise Suite 6 Ease of use Complete interoperability between Borland Optimizeit Request Analyzer, Optimizeit Profiler, Optimizeit Thread Debugger, and Optimizeit Code Coverage: Start any tool from within another; program settings are shared Plug-and-play: no need to recompile code or modify classes before execution Start the tested application directly from the Optimizeit Enterprise Suite start panel Program startup configuration settings saved in setting files for quick startup of frequently tested programs Easily attaches to most application servers, using integration wizards and integration tutorials Detach feature enables quick detach without closing the application server Option to start the tested program from a command line NEW! Enhanced support for headless systems: quick and easy installation and product registration using command-line scripts for Solaris and Linux. NEW! Command-line text-based integration wizards enable easy integration with application servers for systems that do not have a graphical user interface. Source code optional: Optimizeit Enterprise Suite provides performance and reliability information with no need for the source code. Source code access is useful to highlight relevant lines of code NEW! Filter Editor enables certain code items to be filtered out during profiling data capture, using a filter specification file to exclude unwanted code. Filtor Editor can be enforced for CPU, memory, and code coverage profiling data capture to: -Filter out third-party code using modified third-party startup scripts to specify the behavior of the Optimizeit Audit system -Group sets of code into a single backtrace entry -Reduce profiling overhead Start/pause/stop the tested program at any time NEW! Optimizeit Request Analyzer Profile the performance behavior of your J2EE application code across the following J2EE components: Java Database Connectivity (JDBC ), JavaServer Pages (JSP ), Java Naming and Directory Interface (JNDI), Enterprise JavaBeans, and Java Message Service (JMS) containers Improve performance and reliability of J2EE-related application code earlier in development with drill-down performance information for J2EE components Visual interface simplifies the complexity of J2EE application interactions using graphical representation System Dashboard view provides a graphical display of the application time spent in J2EE components and total number of requests. Shows the percentage usage for each server module to quickly detect any major component-level performance issues System Composite view displays all of the J2EE events that have occurred in an application, in real time, in their proper hierarchy. Hierarchy shows the relationship of events in terms of which events spawn others Hotspot view sorts events according to how much time was taken to execute the event Component performance views Java Database Connectivity (JDBC) The Java Database Connectivity (JDBC) view displays activity related to accessing the local or remote database through API calls in JDBC 2

3 Borland Optimizeit Enterprise Suite 6 Identifies the actual SQL statement causing performance problems, making it easier to track down the point in the code where the SQL statement originates JavaServer TM Pages (JSP) The JavaServer Pages (JSP) view displays servlet or JSP invocations in an application for accessing URIs The JSP mode shows how much time is spent responding to a URI request, including invocations of JSP tags or other JSPs. This mode also displays the number of times a JSP or JSP tag is called Java Naming and Directory Interface (JNDI) The Java Naming and Directory Interface (JNDI) view displays the events that are the starting point for accessing many J2EE resources This mode displays calls in real-time and allows users to see precisely which JNDI calls are time consuming and should be cached Enterprise JavaBeans (EJB) The Enterprise JavaBeans (EJB) view displays the overhead associated with EJBs, such as the time spent loading an entity bean and the time spent finding the data Method-level EJB profiling provides drill-down performance information with method-level details capturing time spent and call count Shows the activity surrounding the bean so that the engineer can weed out extraneous bean archiving, bean activation, and bean passivation Java Message Service (JMS) The Java Message Service (JMS) view displays all activity related to queues, topics, and other messaging activity The JMS mode times each API call and shows how much time is spent sending and processing messages NEW! Remote Message Invocation (RMI) NEW! The RMI Details view displays how much time is spent invoking incoming or outgoing remote methods and the number of times a remote method is invoked. NEW! Ability to connect to Optimizeit ServerTrace DataCenter audit system to analyze how efficiently the client application accesses Java objects running in another remote VM in distributed or multi-vm environments Automatic Application Quality Analyzer for J2EE Automatic Application Quality Analyzer preemptively identifies J2EE application performance issues at the code-level at run-time Generates a list of performance-related error conditions and warnings to verify that performance and reliability standards are met Reports on J2EE-specific exceptions that are thrown during testing 3

4 Optimizeit Profiler Ease of use Start the tested application directly from the Optimizeit Profiler start panel Option to pause the tested program immediately after launch to study the program launch phase Export data as ASCII, HTML, or easy-to-parse ASCII to easily print, compare, or archive data Open API allows control of profiling directly from source code Automatic Method Grouping: Simplify profiling data by specifying which methods should be automatically grouped in the call graph Visually isolate critical code with scalable call graph. Select a string allocation and highlight the flow of a method call to see where memory and time are being spent NEW! PDF Reporting: Generate reports in PDF format to capture performance data, including VM information, instance counts, and backtraces for sharing performance data among various teams. High-level performance information High-level performance overview allows developers to understand in real time whether a performance issue is related to CPU, memory, or both Heap graph shows the heap size required and heap size actually used by tested program Thread graph reveals the number of threads running and number of threads actually using CPU Class graph displays the number of classes loaded in the Virtual Machine (VM) Garbage Collector graph tracks garbage collector activity NEW! Automatic Application Quality Analyzer for Java NEW! Supports error prevention and coding standards by automatically detecting VM-level performance bugs NEW! Reports the following errors: Abnormal container growth (StringBuffer, BitSet, ByteArrayOutputStream, Vector, ArrayList, Hashtable, HashMap and all derived classes),- Inefficient StringBuffer use, Garbage collector pauses, Exception Tracing, Abnormal finalizer queue length and resource closed from finalizer NEW! Use with Optimizeit Progress Tracker to track frequency of performance errors Advanced test harness environment Optimizeit Profiler now delivers even lower overhead to create a realistic test environment with minimal intrusion Enhancements for test harness environment provide full control of the CPU Profiler from the audit API Offline profiling: save all performance data in performance snapshots at any given time during a test session; reload data at a later time for further analysis Snapshot timer: Preset a time for one snapshot or a series of snapshots in order to obtain the performance history over a specified test period Ability to run a test and automatically generate a snapshot when the program exits* Progress Tracker measures the impact of performance changes throughout the development lifecycle* Test live applications with offline profiling, combining extremely low overhead profiling and rapid snapshot generation Remote profiling option and the option to attach to application servers facilitate use of Optimizeit Profiler in production environments 4

5 Monitor object allocations in real time with the Heap View Real-time monitoring of object allocations to understand, in real time, how the profiled program uses the Virtual Machine memory Easy-to-read graph shows, in real time, all allocated instances per class with instance count and size Graphical display clearly displays the volume of instances of each class View the total object allocation per count or per size for all classes, with the ability to instantly view the Allocation BackTrace and Instance Reference Graph for the total object allocation* Option for filter class lists based on regular expressions to focus on relevant classes Option to set a reference mark and view the instances allocated since the last mark, to focus on instances allocated by a specific action in the tested program Option to view freed object count to understand performance issues related to the allocation of too many temporary objects Garbage collection controls: Disable garbage collection to study temporary object allocations; enforce garbage collection at any time Track memory leaks Automatic Memory Leak Detector (AMLD) tracks the evolution of memory usage by quickly identifying increases in memory consumption caused by potential memory leaks. New enhancements now make it easier to select an object and immediately access the Reference graph to drill down and highlight the exact line of source code Instance Display Mode tracks all instances of a given class and their incoming and outgoing references, allowing developers to quickly understand why instances are not being garbage-collected Tree display of all references for each instance of a given class and display of the location where these references are allocated Reduced reference graph provides a transitive closure of the full reference graph, to display only references that should be removed in order to free the object for garbage collection Display out-going references to instantly view references kept by each instance of a given class Source Code Viewer automatically highlights relevant lines of code Browses references from root (busy monitors, class static variables, class constants, JNI global and local references, Java threads, and native stacks) to understand exactly the hierarchy between the different objects of the heap Understand object allocations Allocation Backtrace View allows developers to identify the code or part of the program responsible for instance allocations Graph display of all methods involved in instance allocations of a given class with the volume of allocations in percentage Option to reverse display to easily focus on specific methods or lines of code responsible for object allocations Object Size Display automatically computes and displays in real time the amount of memory being consumed by all instances of a class; view and sort by object size to prioritize the object consuming the most memory Listing of methods with the highest allocation count to help identify single methods responsible for excessive allocation Option to refine the display precision from methods to lines of code Automatic highlight of the lines of code relevant to instance allocations with the Source Code Viewer 5

6 CPU profiling Measure pure CPU usage or time usage during a profiling session Display of profiling information per thread and thread groups Color highlight of threads that were busy during profiling session Graph display of all significant methods or lines of code used during the profiling sessions Option to refine the display precision from methods to lines of code Time spent per method or line of code in percentage, absolute time, and invocation count Option to reverse the display to easily focus on specific methods or lines of code responsible for object allocations Hot Spot Display: lists methods where the most time was spent, to help identify bottlenecks due to single methods Automatic highlight of relevant lines of code with the Source Code Viewer CPU profiling: flexibility to match the requirements of test environments Choose between sampling-based profiler and instrumentation-based profiler, according to profiling requirements Sampling-based profiler allows profiling of large amounts of code over a long period of time with millisecond precision while experiencing low overhead Instrumentation-based profiling option allows profiling of small amounts of code with microsecond precision and experience higher overhead Option to control the granularity of sampling-based profiler output by tuning the sampling period Option to control the precision of the instrumentation-based profiler from milliseconds to microseconds Option to filter out instrumentation-based profiler methods that need less time to execute Optimizeit Thread Debugger Real-time view of threads for clear understanding of thread behavior Real-time display of the progress of all threads running within the Virtual Machine Easy-to-read color coding of thread behavior makes it simpler to understand thread issues occurring in real time: thread contentions for a monitor, thread starvation for a resource, and excessive locking View the number of monitors owned by each thread and how many times each thread entered a monitor View which threads are blocked waiting for a monitor and how long they have been blocked View the number of times each thread waited for a monitor and how long each thread waited View the number of times each thread waited for an Input/Output operation and how long it waited Enable/disable view of defunct threads Garbage collector control: force Garbage Collector at any time to easily recreate some thread issues such as those related to finalizer methods Select any thread with a simple mouse click Refine thread selection by selecting a precise time range for a thread, making it easier to understand thread behavior during very precise time periods 6

7 Thread-contention resolution For each thread, understand thread contentions for a monitor with the detailed panels of the Contention View. The Contention View provides the key data needed to understand performance bottlenecks caused by thread contentions Backtrace panel: View the backtrace of method calls for a thread and quickly spot which method is blocking on a monitor Monitor panel: View which monitors are subject to contention Contention detail panel: View a tree of all threads involved in the contention and where the contention took place Automatically highlights lines of code relevant to thread activity with the Source Code Viewer Wait-state monitoring With Waiting View and I/O Waiting View, understand why a thread is not making any progress. Method-call backtrace highlights where threads are waiting for a condition View the cumulated waiting time for each method Automatically highlights lines of code relevant to thread activity with the Source Code Viewer Excessive locking resolution Monitor Enter View reports where a thread enters and holds monitors, allowing developers to understand and correct excessive locking Method-call backtrace highlights the location of entered monitors View the cumulated number of entered monitors for each method Automatically highlights lines of code relevant to thread activity with the Source Code Viewer Resolution of deadlocks Monitor display to understand deadlocks after they occur Graphical view of the relationships between threads and monitors to quickly understand deadlock situations Method-call trace display of thread/monitor relationship to highlight where the thread is blocking and where monitors have been entered Automatically highlights lines of code relevant to thread activity with the Source Code Viewer Prediction of deadlocks and other thread issues Predict deadlocks before they occur with the Monitor Usage Analyzer Start/stop Usage Analyzer recording at any time in a program run Monitor Usage Analyzer generates full list of warnings and errors that may lead to deadlocks and performance bottlenecks such as lock order warnings, lock and wait warnings, and lock and I/O wait warnings. Read a detailed description of each warning to easily take the necessary steps to help avoid deadlocks 7

8 Optimizeit Code Coverage Ease of use Option to filter out packages or classes from tests, in order to focus, for example, on specific parts of code or an application running in server environment Save coverage snapshots and reload them later for further analysis in Optimizeit Code Coverage Generate coverage reports in HTML or ASCII Export coverage data as ASCII, HTML, or importable ASCII to be printed, compared, or archived Find Panel can be accessed from most views to rapidly search and find information Class coverage Real-time class coverage view to quickly see the coverage for each class and identify classes not fully covered Real-time display of all classes and interfaces used by the tested program and real-time percentage of lines covered per class Option to set a reference mark and view the percentage of lines covered since the last mark, to focus on specific parts of the code coverage test Option to filter class lists based on regular expressions to allow focus on relevant classes Options to display in real time the cumulative number of missed lines and cumulative number of missed lines since the last mark Option to display the interfaces that have been loaded by the Virtual Machine Option to display classes and interfaces that have not been loaded Method coverage Method Coverage View displays the methods and lines of code for a selected class that have not been used, allowing developers to modify test plans to cover all areas of the code Method table lists all methods of the class with percentage of code covered for the method, invocation count, and number of missed lines Option to display method invocation count and number of missed lines since the last temporal mark, to focus on specific parts of the test Source Code Viewer highlights the lines of source code that have been covered and lists the number of times each line of code has been executed Source Code Viewer shows lines of code that have never been executed, making it simpler to spot dead code Option to highlight lines of source code executed since the last temporal mark At any time, switch all method coverage information to real time Option to display the signature for each method of the method table Batch-mode coverage Batch-mode support to easily include code coverage in any batch-mode testing process Start coverage from the command line Snapshot Merger merges different runs of the same application from the command line to see coverage of the combined runs 8

9 Open merged snapshots in Optimizeit Code Coverage for easier reading Generate detailed coverage reports in HTML or ASCII directly from the command line Customize reports by defining which code coverage information should be included ENHANCED! Optimizeit Progress Tracker* Select and compare multiple performance snapshots taken from either Optimizeit Profiler or Optimizeit Code Coverage to identify the impact of code changes NEW! Displays comparative results in a graphic as well as in easy-to-read tables as well as with graphical representation NEW! Graph view plots performance variations over time NEW! Supports the Automatic Application Quality Analyzer to track the frequency of performance errors Works with snapshots taken using Optimizeit Suite 4.2 or higher Evaluate comparisons between CPU usage per class and per method Track instance and memory usage by comparing object allocations per class and per method Analyze the comparison differences by sorting by size delta and percent delta NEW! Evaluate changes between Code Coverage snapshots per package Compare code coverage per package, per class and per method NEW! PDF Reporting: Generate reports in PDF format of snapshot comparison tables and performance graphs to share performance progress data * Note: Not available in Optimizeit Suite 5 for IBM AIX and HP-UX. ade in Borland Copyright 2003 Borland Software Corporation. All rights reserved. All Borland brand and product names are trademarks or registered trademarks of Borland Software Corporation in the United States and other countries. Java and all Java-based marks are trademarks or registered trademarks of Sun Microsystems, Inc. in the U.S. and other countries. All other marks are the property of their respective owners