Raja Das

Communication optimizations for irregular scientific computations on distributed memory architectures

Abstract This paper describes a number of optimizations that can be used to support the efficient execution of irregular problems on distributed memory parallel machines. These primitives (1) coordinate interprocessor data movement, (2) manage the storage of, and access to, copies of off-processor data, (3) minimize interprocessor communication requirements, and (4) support a shared name space. We present a detailed performance and scalability analysis of the communication primitives. This performance and scalability analysis is carried out using a workload generator, kernels from real applications, and a large unstructured adaptive application (the molecular dynamics code CHARMM).

Distributed memory compiler design for sparse problems

This paper addresses the issue of compiling concurrent loop nests in the presence of complicated array references and irregularly distributed arrays. Arrays accessed within loops may contain accesses that make it impossible to precisely determine the reference pattern at compile time. This paper proposes a run time support mechanism that is used effectively by a compiler to generate efficient code in these situations. The compiler accepts as input a Fortran 77 program enhanced with specifications for distributing data, and outputs a message passing program that runs on the nodes of a distributed memory machine. The runtime support for the compiler consists of a library of primitives designed to support irregular patterns of distributed array accesses and irregularly distributed array partitions. A variety of performance results on the Intel iPSC/860 are presented. >

Compiler Analysis for Irregular Problems in Fortran D

Many parallel programs require run-time support to implement the communication caused by indirect data references. In previous work, we have developed the inspectorexecutor paradigm to handle these cases. This paper extends that work by developing a dataflow framework to aid in placing the executor communications calls. Our dataflow analysis determines when it is safe to combine communications statements, move them into less frequently executed code regions, or avoid them altogether in favor of reusing data which are already buffered locally.

Run-time and compile-time support for adaptive irregular problems

In adaptive irregular problems, data arrays are accessed via indirection arrays, and data access patterns change during computation. Parallelizing such problems on distributed memory machines requires support for dynamic data partitioning, efficient preprocessing and fast data migration. This paper describes CHAOS, a library of efficient runtime primitives that provides such support. To demonstrate the effectiveness of the runtime support, two adaptive irregular applications have been parallelized using CHAOS primitives: a molecular dynamics code (CHARMM) and a code for simulating gas flows (DSMC). We have also proposed minor extensions to Fortran D which would enable compilers to parallelize irregular for all loops in such adaptive applications by embedding calls to primitives provided by a runtime library. We have implemented our proposed extensions in the Syracuse Fortran 90D/HPF prototype compiler, and have used the compiler to parallelize kernels from two adaptive applications.<<ETX>>

Runtime and language support for compiling adaptive irregular programs on distributed-memory machines

In many scientific applications, arrays containing data are indirectly indexed through indirection arrays. Such scientific applications are called irregular programs and are a distinct class of applications that require special techniques for parallelization.

Jovian: a framework for optimizing parallel I/O

There has been a great deal of recent interest in parallel I/O. We discuss the design and implementation of the Jovian library, which is intended to optimize the I/O performance of multiprocessor architectures that include multiple disks or disk arrays. We also present preliminary performance measurements from benchmarking the Jovian I/O library on the IBM SP1 distributed memory parallel machine for two application templates.<<ETX>>

Interprocedural partial redundancy elimination and its application to distributed memory compilation

Partial Redundancy Elimination (PRE) is a general scheme for suppressing partial redundancies which encompasses traditional optimizations like loop invariant code motion and redundant code elimination. In this paper we address the problem of performing this optimization interprocedurally. We use interprocedural partial redundancy elimination for placement of communication and communication preprocessing statements while compiling for distributed memory parallel machines.

Slicing Analysis and Indirect Accesses to Distributed Arrays

An increasing fraction of the applications targeted by parallel computers makes heavy use of indirection arrays for indexing data arrays. Such irregular access patterns make it difficult for a compiler to generate efficient parallel code. Previously developed techniques addressing this problem are limited in that they are only applicable for a single level of indirection. However, many codes using sparse data structures access their data through multiple levels of indirection.

Distributed memory compiler methods for irregular problems—data copy reuse and runtime partitioning

Abstract This paper outlines two methods which we believe will play an important role in any distributed memory compiler able to handle sparse and unstructured problems. We describe how to link runtime partitioners to distributed memory compilers. In our scheme, programmers can implicitly specify how data and loop iterations are to be distributed between processors. This insulates users from having to deal explicitly with potentially complex algorithms that carry out work and data partitioning. We also describe a viable mechanism for tracking and reusing copies of off-processor data. In many programs, several loops access the same off-processor memory locations. As long as it can be verified that the values assigned to off-processor memory locations remain unmodified, we show that we can effectively reuse stored off-processor data. We present experimental data from a 3-D unstructured Euler solver run on an iPSC/860 to demonstrate the usefulness of our methods.

Implementation of a parallel unstructured Euler solver on shared and distributed memory architectures

An efficient three-dimensional unstructured Euler solver has been parallelized on a Cray Y-MP C90 shared memory computer and on an Intel Touchstone Delta distributed memory computer. Both machines yield comparable performance rates. However, the availability of sophisticated software tools enabled the parallelization of EUL3D on the shared memory vector/parallel CRAY Y-MP C90 with minimal user input. On the other hand, the implementation on the distributed memory massively parallel architecture of the Intel Touchstone Delta machine is considerably more involved. As massively parallel software tools become more mature, the task of developing or porting software to such machines should diminish. It has also been shown that with todays supercomputers, and with efficient codes such as EUL3D, the aerodynamic characteristics of complex vehicles can be computed in a matter of minutes, making design use feasible.<<ETX>>