uu.seUppsala University Publications
Change search
ReferencesLink to record
Permanent link

Direct link
Performance modelling for parallel PDE solvers on NUMA-systems
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Scientific Computing. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Numerical Analysis. (Software Aspects of High-Performance Computing)
2006 (English)Report (Other academic)
Place, publisher, year, edition, pages
Technical report / Department of Information Technology, Uppsala University, ISSN 1404-3203 ; 2006-041
National Category
Computer Science Computational Mathematics
URN: urn:nbn:se:uu:diva-81930OAI: oai:DiVA.org:uu-81930DiVA: diva2:109845
Available from: 2008-02-19 Created: 2008-02-19 Last updated: 2011-11-18Bibliographically approved
In thesis
1. Multithreaded PDE Solvers on Non-Uniform Memory Architectures
Open this publication in new window or tab >>Multithreaded PDE Solvers on Non-Uniform Memory Architectures
2006 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

A trend in parallel computer architecture is that systems with a large shared memory are becoming more and more popular. A shared memory system can be either a uniform memory architecture (UMA) or a cache coherent non-uniform memory architecture (cc-NUMA).

In the present thesis, the performance of parallel PDE solvers on cc-NUMA computers is studied. In particular, we consider the shared namespace programming model, represented by OpenMP. Since the main memory is physically, or geographically distributed over several multi-processor nodes, the latency for local memory accesses is smaller than for remote accesses. Therefore, the geographical locality of the data becomes important.

The focus of the present thesis is to study multithreaded PDE solvers on cc-NUMA systems, in particular their memory access pattern with respect to geographical locality. The questions posed are: (1) How large is the influence on performance of the non-uniformity of the memory system? (2) How should a program be written in order to reduce this influence? (3) Is it possible to introduce optimizations in the computer system for this purpose?

The main conclusion is that geographical locality is important for performance on cc-NUMA systems. This is shown experimentally for a broad range of PDE solvers as well as theoretically using a model involving characteristics of computer systems and applications.

Geographical locality can be achieved through migration directives that are inserted by the programmer or — possibly in the future — automatically by the compiler. On some systems, it can also be accomplished by means of transparent, hardware initiated migration and replication. However, a necessary condition that must be fulfilled if migration is to be effective is that the memory access pattern must not be "speckled", i.e. as few threads as possible shall make accesses to each memory page.

We also conclude that OpenMP is competitive with MPI on cc-NUMA systems if care is taken to get a favourable data distribution.

Place, publisher, year, edition, pages
Uppsala: Acta Universitatis Upsaliensis, 2006. 33 p.
Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology, ISSN 1651-6214 ; 224
PDE solver, high-performance, NUMA, UMA, OpenMP, MPI, data migration, data replication, thread scheduling, data affinity
National Category
Software Engineering
Research subject
Scientific Computing
urn:nbn:se:uu:diva-7149 (URN)91-554-6656-7 (ISBN)
Public defence
2006-10-20, Room 2446, Polacksbacken, Lägerhyddsvägen 2D, Uppsala, 10:15 (English)
Available from: 2006-09-28 Created: 2006-09-28 Last updated: 2011-10-26Bibliographically approved

Open Access in DiVA

No full text

Other links


Search in DiVA

By author/editor
Nordén, Markus
By organisation
Division of Scientific ComputingNumerical Analysis
Computer ScienceComputational Mathematics

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 211 hits
ReferencesLink to record
Permanent link

Direct link