Performance and Debugging Tools¶
NERSC provides many popular debugging and profiling tools. Some of them are general-purpose tools and others are geared toward more specific tasks.
A quick guideline on when to use which debugging tool is as follows:
- DDT and TotalView: general purpose parallel debuggers allowing users to interactively control the pace of execution of a program using a graphical user interface
- gdb: a serial command line mode debugger; can be useful in quickly examining core files to see where the code crashed (DDT and TotalView can be used for this purpose as well)
- STAT: used for obtaining call backtraces for all parallel tasks from a live parallel application and displaying a call tree graphically, showing where each task is executing; useful in debugging a hung application
- ATP: used for generating call backtraces for all parallel tasks when a code crashes; useful in debugging a hung application; can be a good starting point if a code crashes with little hint left behind
- CCDB and lgdb: run two versions of a code (e.g., one working version and an incorrect version, or a code run with two different numbers of tasks) side by side to find out where the two runs start to generate diverging results
- Valgrind: a suite of debugging and profiling tools; the best known tool is memcheck which can detect memory errors or memory leaks; other tools include cache profiling, heap memory profiling tools, and more
- Intel Inspector: a memory and threading error-checking tool for users developing serial and multithreaded applications on Windows and Linux operating systems
Two "getting started" tutorials on some debugging tools are:
A quick guideline for performance analysis tools is as follows:
- IPM: a low-overhead easy-to-use tool for getting hardware counters data, MPI function timings, and memory usage
- CrayPat: a suite of sophisticated Cray tools for a more detailed performance analysis which can show routine-based hardware counter data, MPI message statistics, I/O statistics, etc; in addition to getting performance data deduced from a sampling method, tracing of certain routines (or library routines) can be performed for better understanding of performance statistics associated with the selected routines
- MAP: a sampling tool for performance metrics; time series of the collected data for the entire run of the code is displayed graphically, and the source code lines are annotated with performance metrics
- Intel VTune Amplifier XE: a GUI-based tool that can find performance bottlenecks
A "getting started" tutorial on some performance analysis is:
For more information about how to use a tool, click on the relevant item below:
TotalView: TotalView, from Rogue Wave Software, is a parallel debugging tool that can be run with up to 512 processors. It provides an X Windows-based Graphical User Interface and a command line interface. Read more here.
STAT and ATP: STAT (the Stack Trace Analysis Tool) is a highly scalable, lightweight tool that gathers and merges stack traces from all of the processes of a parallel application. ATP (Abnormal Termination Processing) automatically runs STAT when the code crashes. Read more here.
CCDB and lgdb: lgdb (Cray Line Mode Parallel Debugger) is a GDB-based parallel debugger, developed by Cray. It allows programmers to either launch an application or attach to an already-running application that was launched with
srun in order to debug the parallel code in command-line mode. Read more here.
Valgrind: The Valgrind tool suite provides several debugging and profiling tools that can help make your programs faster and more correct. The most popular tool is Memcheck, which can detect many memory-related errors that are common in C and C++ programs. Read more here.
IPM: IPM is a portable profiling infrastructure which provides a high-level report on the execution of a parallel job. IPM reports hardware counter data, MPI function timings, and memory usage. Read more here.
Application Performance Snapshot (APS): Application Performance Snapshot (APS) is a lightweight open source profiling tool developed by the Intel VTune developers. Use APS for a quick view into a shared memory or MPI application's use of available hardware (CPU, FPU, and memory). APS analyzes your application's time spent in MPI, MPI and OpenMP imbalance, memory access efficiency, FPU usage, and I/O and memory footprint. Read more here.
LIKWID: LIKWID ("Like I Knew What I'm Doing") is a lightweight suite of command line utilities. By reading the the MSR (Model Specific Register) device files, it renders reports for various performance metrics such as FLOPS, bandwidth, load to store ratio, and energy. Read more here. Read more here.
Intel Advisor: Intel Advisor provides two workflows to help ensure that Fortran, C and C++ applications can make the most of today's processors: vectorization advisor and threading advisor. Read more here.
Intel Trace Analyzer and Collector: Intel Trace Analyzer and Collector (ITAC) are two tools used for analyzing MPI behavior in parallel applications. ITAC identifies MPI load imbalance and communication hotspots in order to help developers optimize MPI parallelization and minimize communication and synchronization in their applications. Read more here.