4. Software Architecture

This chapter focuses on describing the WRAPPER environment within which both the core numerics and the pluggable packages operate. The description presented here is intended to be a detailed exposition and contains significant background material, as well as advanced details on working with the WRAPPER. The tutorial sections of this manual (see sections 3.7 and 5.3) contain more succinct, step-by-step instructions on running basic numerical experiments, of varous types, both sequentially and in parallel. For many projects simply starting from an example code and adapting it to suit a particular situation will be all that is required. The first part of this chapter discusses the MITgcm architecture at an abstract level. In the second part of the chapter we described practical details of the MITgcm implementation and of current tools and operating system features that are employed.

• 4.1 Overall architectural goals
• 4.2 WRAPPER
  • 4.2.1 Target hardware
  • 4.2.2 Supporting hardware neutrality
  • 4.2.3 WRAPPER machine model
  • 4.2.4 Machine model parallelism
  • 4.2.5 Communication mechanisms
  • 4.2.6 Shared memory communication
  • 4.2.7 Distributed memory communication
  • 4.2.8 Communication primitives
  • 4.2.9 Memory architecture
  • 4.2.10 Summary
  
  
• 4.3 Using the WRAPPER
  • 4.3.1 Specifying a domain decomposition
  • 4.3.2 Starting the code
  • 4.3.3 Controlling communication
  
  
• 4.4 MITgcm execution under WRAPPER
  • 4.4.1 Annotated call tree for MITgcm and WRAPPER
  • 4.4.2 Measuring and Characterizing Performance
  • 4.4.3 Estimating Resource Requirements

Copyright © 2002 Massachusetts Institute of Technology