یکی از نیازهای مهم در کلاسترهای محاسباتی و سیستمهای پردازش موازی، ارزیابی میزان کارایی سخت افزار و نرم افزار محاسباتی طراحی شده می باشد. در این مطلب یک Benchmark از ناسا معرفی شده است که ترکیبی از اجرای مسائل پیچیده محاسباتی با ورودیهای مختلف در تعدادی از رشته های تخصصی ریاضی، مکانیک سیالات و .. می باشد. این محاسبات با چارچوبهای مختلفی از قبیل OpenMP, MPI و زبانهای برنامه نویسی C و Fortran نوشته شده است.

NAS PARALLEL BENCHMARKS

The NAS Parallel Benchmarks (NPB) are a small set of programs designed to help evaluate the performance of parallel supercomputers. The benchmarks are derived from computational fluid dynamics (CFD) applications and consist of five kernels and three pseudo-applications in the original "pencil-and-paper" specification (NPB 1). The benchmark suite has been extended to include new benchmarks for unstructured adaptive mesh, parallel I/O, multi-zone applications, and computational grids.  Problem sizes in NPB are predefined and indicated as different classes. Reference implementations of NPB are available in commonly-used programming models like MPI and OpenMP (NPB 2 and NPB 3).

BENCHMARK SPECIFICATIONS

The original eight benchmarks specified in NPB 1 mimic the computation and data movement in CFD applications:

• five kernels
  • IS - Integer Sort, random memory access
  • EP - Embarrassingly Parallel
  • CG - Conjugate Gradient, irregular memory access and communication
  • MG - Multi-Grid on a sequence of meshes, long- and short-distance communication, memory intensive
  • FT - discrete 3D fast Fourier Transform, all-to-all communication
• three pseudo applications
  • BT - Block Tri-diagonal solver
  • SP - Scalar Penta-diagonal solver
  • LU - Lower-Upper Gauss-Seidel solver

Multi-zone versions of NPB (NPB-MZ) are designed to exploit multiple levels of parallelism in applications and to test the effectiveness of multi-level and hybrid parallelization paradigms and tools.  There are three types of benchmark problems derived from single-zone pseudo applications of NPB:

• BT-MZ - uneven-size zones within a problem class, increased number of zones as problem class grows
• SP-MZ - even-size zones within a problem class, increased number of zones as problem class grows
• LU-MZ - even-size zones within a problem class, a fixed number of zones for all problem classes

Benchmarks for unstructured computation, parallel I/O, and data movement

• UA - Unstructured Adaptive mesh, dynamic and irregular memory access
• BT-IO - test of different parallel I/O techniques
• DC - Data Cube
• DT - Data Traffic

GridNPB is designed specifically to rate the performance of computational grids. Each of the four benchmarks in the set consists of a collection of communicating tasks derived from the NPB. They symbolize distributed applications typically run on grids.

• ED - Embarrassingly Distributed
• HC - Helical Chain
• VP - Visualization Pipeline
• MB - Mixed Bag

DOCUMENTS ON BENCHMARK SPECIFICATIONS:

• NAS Parallel Benchmarks, RNR-94-007 (PDF-425KB) for IS, EP, CG, MG, FT, BT, SP, LU
• NAS Parallel Benchmarks I/O Version 2.4, NAS-03-002 (PDF-57KB) for BT-IO
• Unstructured Adaptive NAS Parallel Benchmark, NAS-04-006 (PDF-184KB) for UA
• NAS Parallel Benchmarks, Multi-Zone Versions, NAS-03-010 (PDF-128KB) for BT-MZ, SP-MZ, LU-MZ
• Arithmetic Data Cube as a Data Intensive Benchmark, NAS-03-005 (PDF-65KB) for DC
• NAS Grid Benchmarks Version 1.0, NAS-02-005 (PDF-352KB) for GridNPB

BENCHMARK CLASSES

• Class S: small for quick test purposes
• Class W: workstation size (a 90's workstation; now likely too small)
• Classes A, B, C: standard test problems; ~4X size increase going from one class to the next
• Classes D, E, F: large test problems; ~16X size increase from each of the previous classes

Details of NPB Problem Sizes and Parameters.

REFERENCE IMPLEMENTATIONS

Vendors and others implement the detailed specifications in the NPB 1 report, using algorithms and programming models appropriate to their different machines. NPB 1 implementations are generally proprietary and are not distributed by NAS.

A set of reference implementations of the NPB specifications has been written and distributed by NAS as NPB 2 and NPB 3. These source-code implementations are intended to be run with little or no tuning, and approximate the performance a typical user can expect to obtain for a portable parallel program. NPB 2 contains MPI-based source code implementations of the original eight benchmarks, and NPB 3 has included new benchmarks and problem classes as well as implementations using other programming models.  The latest release is NPB 3.3.1.