Jeffrey M. Squyres

The Lam/Mpi Checkpoint/Restart Framework: System-Initiated Checkpointing

As high performance clusters continue to grow in size and popularity, issues of fault tolerance and reliability are becoming limiting factors on application scalability. To address these issues, we present the design and implementation of a system for providing coordinated checkpointing and rollback recovery for MPI-based parallel applications. Our approach integrates the Berkeley Lab BLCR kernel-level process checkpoint system with the LAM implementation of MPI through a defined checkpoint/restart interface. Checkpointing is transparent to the application, allowing the system to be used for cluster maintenance and scheduling reasons as well as for fault tolerance. Experimental results show negligible communication performance impact due to the incorporation of the checkpoint support capabilities into LAM/MPI.

Object Oriented MPI (OOMPI): a class library for the Message Passing Interface

Using the Message Passing Interface (MPI) in C++ has been difficult up to this point, because of the lack of suitable C++ bindings and C++ class libraries. The existing MPI standard provides language bindings only for C and Fortran 77, precluding their direct use in object-oriented programming. Even the proposed C++ bindings in MPI-2 are at a fairly low-level and are not directly suitable for object-oriented programming. In this paper, we present the requirements, analysis and design for Object-Oriented MPI (OOMPI), a C++ class library for MPI. Although the OOMPI class library is specified in C++, in some sense the specification is a generic one that uses C++ as the program description language. Thus, the OOMPI specification can also be considered as a generic object-oriented class library specification which can thus also form the basis for MPI class libraries in other object-oriented languages.

The Design and Evolution of the MPI-2 C++ Interface

The original specification for the Message Passing Interface (MPI) included language bindings for C and Fortran 77. C++ programs that used MPI were thus required to use the C bindings. With MPI-2, a C++ interface for all of MPI is specified. In this paper, we describe the design of the C++ interface for MPI and provide some of the history and motivations behind the design decisions.

Object-oriented analysis and design of the Message Passing Interface

The major contribution of this paper is the application of modern analysis techniques to the important Message Passing Interface standard, work done in order to obtain information useful in designing both application programmer interfaces for object‐oriented languages, and message passing systems. Recognition of ‘Design Patterns’ within MPI is an important discernment of this work. A further contribution is a comparative discussion of the design and evolution of three actual object‐oriented designs for the Message Passing Interface ( MPI‐1SF ) application programmer interface (API), two of which have influenced the standardization of C++ explicit parallel programming with MPI‐2, and which strongly indicate the value of a priori object‐oriented design and analysis of such APIs. Knowledge of design patterns is assumed herein.

Waveform iterative methods for parallel solution of initial value problems

The traditional approach for computing the solution to large systems of ordinary differential or differential-algebraic equations typically includes discretization in time with an implicit integration formula. The primary opportunity for parallelization is therefore limited to the linear system solution that is performed at each timestep. Waveform techniques, on the other hand, decompose the problem at the equation level and solve for different components of the system independently, using previous iterates from other processors as inputs. This approach is particularly well-suited for message-passing computing environments, especially those with high communication latency because synchronization and communication take place infrequently and communication consists of large packets of information. We present an MPI-based implementation of a waveform relaxation-based semiconductor device simulation program and provide experimental results using this program to solve the time dependent semiconductor drift-diffusion equations on a cluster of workstations.<<ETX>>

Open MPI community meeting

The Open MPI Project is a growing community surrounding an open source implementation of the Message Passing Interface, developed in a collaboration between research institutions, HPC vendors, and US national laboratories. Open MPI has evolved into a high performance, scalable MPI implementation, providing a highly modular architecture that not only adapts to a wide variety of environments, but also uniquely lends itself to HPC research.The meeting will consist of three parts:1. Members of the Open MPI core development team will be presenting the current status of Open MPI. 2. Discuss possible future directions for Open MPI, actively soliciting feedback from real-world MPI users and ISVs with MPI-based products. 3. Active discussion about how HPC researchers can leverage the Open MPI project for their own work, and how MPI users can obtain the best performance out of real-world applications.