Jarmo Rantakokko

Jarmo Rantakokko

Auto-generated profile, based on publicly available information

Jarmo Rantakokko

Uppsala University

Research Focus & Metrics

Academic Career Length

25Years (2002-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 September 2007 in International Journal of Parallel Programming
Verified
Journal Article
Closed access
PDF file
MNMarkus Nordén
Markus Nordén
HLHenrik Löf
Henrik Löf
JRJarmo Rantakokko
Jarmo Rantakokko
SHSverker Holmgren
Sverker Holmgren

1 April 2002 in Parallel Algorithms and Applications
Verified
Journal Article
Closed access
MNMarkus Nordén
Markus Nordén
JRJarmo Rantakokko
Jarmo Rantakokko
DWDan Wallin
Dan Wallin
SHSverker Holmgren
Sverker Holmgren
The performance of shared-memory (OpenMP) implementa- tions of three dierent PDE solver kernels representing finite dierence methods, finite volume methods, and spectral methods has been investi- gated. The experiments have been performed on a self-optimizing NUMA system, the Sun Orange prototype, using dierent data placement and thread scheduling strategies. The results show that correct data place- ment is very important for the performance for all solvers. However, the Orange system has a unique capababilty of automatically changing the data distribution at run time through both migration and replication of data. For reasonable large PDE problems, we find that the time to do this is negligible compared to the total solve time. Also, the performance after the migration and replication process has reached steady-state is the same as what is achieved if data is optimally placed at the begin- ning of the execution using hand tuning. This shows that, for the ap- plication studied, the self-optimizing features are successful, and shared memory code without explicit data distribution directives yields good performance.

1 October 2006 in International Journal of Parallel, Emergent and Distributed Systems
Verified
Journal Article
Closed access
HLHenrik Löf
Henrik Löf
JRJarmo Rantakokko
Jarmo Rantakokko
OpenMP is an architecture-independent language for programming in the shared memory model. OpenMP is designed to be simple and powerful in terms of programming abstractions. Unfortunately, the architecture-independent abstractions sometimes come with the price of low parallel performance. This is especially true for applications with an unstructured data access pattern running on distributed shared memory systems (DSM). Here, proper data distribution and algorithmic optimizations play a vital role for performance. In this article, we have investigated ways of improving the performance of an industrial class conjugate gradient (CG) solver, implemented in OpenMP running on two types of shared memory systems. We have evaluated bandwidth minimization, graph partitioning and reformulations of the original algorithm reducing global barriers. By a detailed analysis of barrier time and memory system performance, we found that bandwidth minimization is the most important optimization reducing both L2 misses and remote memory accesses. On a uniform memory system, we get perfect scaling. On a NUMA system, the performance is significantly improved with the algorithmic optimizations leaving the system dependent global reduction operations as a bottleneck.