DDT on Franklin

Distributed Debugging Tool (DDT) from Allinea Software is the new parallel debugger for Franklin.

Contents

• Introduction
• Loading the DDT module
• Compiling code to run with DDT
• Starting a Job with DDT
• Basic Functionality
• Useful New DDT Features

Introduction

DDT is a parallel debugger which can be run on Franklin with up to 1024 processors. Totalview users will find DDT has very similar functionality and an intuitive user interface. All of the primary parallel debugging features from Totalview are available with DDT. Please let us know how you like DDT and report any questions or problems to consult@nersc.gov

The Allinea DDT web page and users guide is a good resource for learning more about some of the advanced DDT features.

Loading the DDT Module

To use DDT at NERSC, first load the DDT module to set the correct environment settings with the following command:

Franklin% module load ddt

Compiling Code to run with DDT

In order to use DDT, code must be compiled with the -g option. We also recommend that you do not run with optimization turned on, flags such as -fast.

A Fortran example:

Franklin% ftn -g -o testDDT_ex testDDT.f

Franklin% cc -g -o testDDT_ex testDDT.c

Starting a Job with DDT

Be sure to log into Franklin with an X window forwarding enabled. This could mean using the -X or -Y option to ssh. The -Y option often works better for Mac OSX.

% ssh -X username@Franklin.nersc.gov

Franklin% ddt testDDT_ex

Basic Debugging Functionality

Useful New DDT Features

Process Groups

With DDT, the user can easily change the debugger to focus on a single process or group of processes. If Focus on Processor is chosen, then stepping through the code, setting a breakpoint etc will occur only for a given processor. If Focus on Group is chosen then the entire group of processors will advance when stepping forward in a program and a breakpoint will be set for all processors in a group. In addition, a user can create new sub-groups of processors by right clicking in the Process Group Window and then dragging the desired processors to the group. Groups can also be created more efficiently using sub-groups from the Parallel Stack View described below. The below image shows 3 different groups of processors, the default All group, a group with only a single master processor Root and a group with the remaining Worker processors.

Parallel Stack View

A new feature which should help users debug at high concurrencies is DDT's Parallel Stack View window which allows the user to see the position of all processors in a code at the same time from the main window. A program is displayed as a branching tree with the number and location of each processor at each point. Instead of clicking through windows to determine where each processor has stopped, the Parallel Stack View presents a quick overview which easily allows users to identify stray processes. Users can also create sub-groups of processors from a branch of the tree by right clicking on the branch. A new group will appear in the Process Group Window at the top of the GUI.

Memory Debugging

DDT has a memory debugging tool which shows heap memory usage across processors. In order to use memory debugging, a user must link with the following option -Bstaticddt. Memory debugging is fully enabled for C codes and more functionality should be available for Fortran codes soon.

Franklin% cc -g testDDT.c -Bstaticddt