Logo: to the web site of Uppsala University

uu.sePublikasjoner fra Uppsala universitet
Endre søk
Link to record
Permanent link

Direct link
Nordén, Markus
Publikasjoner (10 av 11) Visa alla publikasjoner
Nordén, M., Löf, H., Rantakokko, J. & Holmgren, S. (2007). Dynamic data migration for structured AMR solvers. International journal of parallel programming, 35, 477-491
Åpne denne publikasjonen i ny fane eller vindu >>Dynamic data migration for structured AMR solvers
2007 (engelsk)Inngår i: International journal of parallel programming, ISSN 0885-7458, E-ISSN 1573-7640, Vol. 35, s. 477-491Artikkel i tidsskrift (Fagfellevurdert) Published
HSV kategori
Identifikatorer
urn:nbn:se:uu:diva-11396 (URN)10.1007/s10766-007-0056-z (DOI)000249405600004 ()
Tilgjengelig fra: 2007-09-11 Laget: 2007-09-11 Sist oppdatert: 2018-01-12bibliografisk kontrollert
Nordén, M., Löf, H., Rantakokko, J. & Holmgren, S. (2006). Geographical locality and dynamic data migration for OpenMP implementations of adaptive PDE solvers.
Åpne denne publikasjonen i ny fane eller vindu >>Geographical locality and dynamic data migration for OpenMP implementations of adaptive PDE solvers
2006 (engelsk)Rapport (Annet vitenskapelig)
Abstract [en]

On cc-NUMA multi-processors, the non-uniformity of main memory latencies motivates the need for co-location of threads and data. We call this special form of data locality, geographical locality. In this article, we study the performance of a parallel PDE solver with adaptive mesh refinement. The solver is parallelized using OpenMP and the adaptive mesh refinement makes dynamic load balancing necessary. Due to the dynamically changing memory access pattern caused by the runtime adaption, it is a challenging task to achieve a high degree of geographical locality.

The main conclusions of the study are: (1) that geographical locality is very important for the performance of the solver, (2) that the performance can be improved significantly using dynamic page migration of misplaced data, (3) that a migrate-on-next-touch directive works well whereas the first-touch strategy is less advantageous for programs exhibiting a dynamically changing memory access patterns, and (4) that the overhead for such migration is low compared to the total execution time.

Serie
Technical report / Department of Information Technology, Uppsala University, ISSN 1404-3203 ; 2006-038
HSV kategori
Identifikatorer
urn:nbn:se:uu:diva-81928 (URN)
Merknad

Note: To appear in Proceedings of the 2:nd International Workshop on OpenMP (IWOMP)

Tilgjengelig fra: 2008-02-15 Laget: 2008-02-15 Sist oppdatert: 2024-05-31bibliografisk kontrollert
Nordén, M. (2006). Multithreaded PDE Solvers on Non-Uniform Memory Architectures. (Doctoral dissertation). Uppsala: Acta Universitatis Upsaliensis
Åpne denne publikasjonen i ny fane eller vindu >>Multithreaded PDE Solvers on Non-Uniform Memory Architectures
2006 (engelsk)Doktoravhandling, med artikler (Annet vitenskapelig)
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.

sted, utgiver, år, opplag, sider
Uppsala: Acta Universitatis Upsaliensis, 2006. s. 33
Serie
Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology, ISSN 1651-6214 ; 224
Emneord
PDE solver, high-performance, NUMA, UMA, OpenMP, MPI, data migration, data replication, thread scheduling, data affinity
HSV kategori
Forskningsprogram
Beräkningsvetenskap
Identifikatorer
urn:nbn:se:uu:diva-7149 (URN)91-554-6656-7 (ISBN)
Disputas
2006-10-20, Room 2446, Polacksbacken, Lägerhyddsvägen 2D, Uppsala, 10:15 (engelsk)
Opponent
Veileder
Tilgjengelig fra: 2006-09-28 Laget: 2006-09-28 Sist oppdatert: 2022-03-11bibliografisk kontrollert
Nordén, M., Holmgren, S. & Thuné, M. (2006). OpenMP versus MPI for PDE solvers based on regular sparse numerical operators. Future generations computer systems, 22, 194-203
Åpne denne publikasjonen i ny fane eller vindu >>OpenMP versus MPI for PDE solvers based on regular sparse numerical operators
2006 (engelsk)Inngår i: Future generations computer systems, ISSN 0167-739X, E-ISSN 1872-7115, Vol. 22, s. 194-203Artikkel i tidsskrift (Fagfellevurdert) Published
HSV kategori
Identifikatorer
urn:nbn:se:uu:diva-47210 (URN)10.1016/j.future.2003.09.004 (DOI)000234408800016 ()
Tilgjengelig fra: 2006-05-23 Laget: 2006-05-23 Sist oppdatert: 2018-01-11bibliografisk kontrollert
Nordén, M. (2006). Performance modelling for parallel PDE solvers on NUMA-systems.
Åpne denne publikasjonen i ny fane eller vindu >>Performance modelling for parallel PDE solvers on NUMA-systems
2006 (engelsk)Rapport (Annet vitenskapelig)
Abstract [en]

A detailed model of the memory performance of a PDE solver running on a NUMA-system is set up. Due to the complexity of modern computers, such a detailed model inevitably is very complicated. Therefore, approximations are introduced that simplify the model and allows NUMA-systems and PDE solvers to be described conveniently.

Using the simplified model, it is shown that PDE solvers using ordered local methods can be made very unsensitive to high NUMA-ratios, allowing them to scale well on virtually any NUMA-system.

PDE solvers using unordered local methods, semiglobal methods or global methods are more sensitive to high NUMA-ratios and require special techniques in order to scale well beyond a single locality group.

Nevertheless, the potential performance gain of improving the data distribution on a NUMA-system can be considerable for all kinds of PDE solvers studied.

Serie
Technical report / Department of Information Technology, Uppsala University, ISSN 1404-3203 ; 2006-041
HSV kategori
Identifikatorer
urn:nbn:se:uu:diva-81930 (URN)
Tilgjengelig fra: 2008-02-19 Laget: 2008-02-19 Sist oppdatert: 2024-05-31bibliografisk kontrollert
Löf, H., Nordén, M. & Holmgren, S. (2004). Improving Geographical Locality of Data for Shared Memory Implementations of PDE Solvers. In: Computational Science – ICCS 2004 (pp. 9-16). Berlin: Springer-Verlag
Åpne denne publikasjonen i ny fane eller vindu >>Improving Geographical Locality of Data for Shared Memory Implementations of PDE Solvers
2004 (engelsk)Inngår i: Computational Science – ICCS 2004, Berlin: Springer-Verlag , 2004, s. 9-16Konferansepaper, Publicerat paper (Fagfellevurdert)
sted, utgiver, år, opplag, sider
Berlin: Springer-Verlag, 2004
Serie
Lecture Notes in Computer Science ; 3037
HSV kategori
Identifikatorer
urn:nbn:se:uu:diva-71098 (URN)10.1007/b97988 (DOI)
Tilgjengelig fra: 2007-03-11 Laget: 2007-03-11 Sist oppdatert: 2018-01-10bibliografisk kontrollert
Löf, H., Nordén, M. & Holmgren, S. (2004). Improving geographical locality of data for shared memory implementations of PDE solvers.
Åpne denne publikasjonen i ny fane eller vindu >>Improving geographical locality of data for shared memory implementations of PDE solvers
2004 (engelsk)Rapport (Annet vitenskapelig)
Abstract [en]

On cc-NUMA multi-processors, the non-uniformity of main memory latencies motivates the need for co-location of threads and data. We call this special form of data locality, geographical locality, as the non-uniformity is a consequence of the physical distance between the cc-NUMA nodes. In this article, we compare the well established method of exploiting the first-touch strategy using parallel initialization of data to an application-initiated page migration strategy as means of increasing the geographical locality for a set of important scientific applications.

Four PDE solvers parallelized using OpenMP are studied; two standard NAS NPB3.0-OMP benchmarks and two kernels from industrial applications. The solvers employ both structured and unstructured computational grids. The main conclusions of the study are: (1) that geographical locality is important for the performance of the applications, (2) that application-initiated migration outperforms the first-touch scheme in almost all cases, and in some cases even results in performance which is close to what is obtained if all threads and data are allocated on a single node. We also suggest that such an application-initiated migration could be made fully transparent by letting the OpenMP compiler invoke it automatically.

Serie
Technical report / Department of Information Technology, Uppsala University, ISSN 1404-3203 ; 2004-006
HSV kategori
Identifikatorer
urn:nbn:se:uu:diva-66907 (URN)
Tilgjengelig fra: 2006-12-05 Laget: 2006-12-05 Sist oppdatert: 2024-05-31bibliografisk kontrollert
Nordén, M. (2004). Parallel PDE Solvers on cc-NUMA Systems. (Licentiate dissertation). Uppsala University
Åpne denne publikasjonen i ny fane eller vindu >>Parallel PDE Solvers on cc-NUMA Systems
2004 (engelsk)Licentiatavhandling, med artikler (Annet vitenskapelig)
Abstract [en]

The current trend in parallel computers 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 questions posed in this thesis 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?

Most of the application codes studied address the Euler equations using a finite difference method and a finite volume method respectively and are parallelized with OpenMP. Comparisons are made with an alternative implementation using MPI and with PDE solvers implemented with OpenMP that solve other equations using different numerical methods.

The main conclusion is that geographical locality is important for performance on cc-NUMA systems. This can be achieved through self optimization provided in the system or through migrate-on-next-touch directives that could be inserted automatically by the compiler.

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

sted, utgiver, år, opplag, sider
Uppsala University, 2004
Serie
IT licentiate theses / Uppsala University, Department of Information Technology, ISSN 1404-5117 ; 2004-002
HSV kategori
Forskningsprogram
Beräkningsvetenskap
Identifikatorer
urn:nbn:se:uu:diva-86307 (URN)
Veileder
Tilgjengelig fra: 2004-03-26 Laget: 2006-05-14 Sist oppdatert: 2018-01-13bibliografisk kontrollert
Nordén, M., Holmgren, S. & Thuné, M. (2002). OpenMP versus MPI for PDE solvers based on regular sparse numerical operators. In: Computational Science – ICCS 2002 (pp. 681-690). Berlin: Springer-Verlag
Åpne denne publikasjonen i ny fane eller vindu >>OpenMP versus MPI for PDE solvers based on regular sparse numerical operators
2002 (engelsk)Inngår i: Computational Science – ICCS 2002, Berlin: Springer-Verlag , 2002, s. 681-690Konferansepaper, Publicerat paper (Annet vitenskapelig)
sted, utgiver, år, opplag, sider
Berlin: Springer-Verlag, 2002
Serie
Lecture Notes in Computer Science ; 2331
HSV kategori
Identifikatorer
urn:nbn:se:uu:diva-44296 (URN)
Tilgjengelig fra: 2006-05-17 Laget: 2006-05-17 Sist oppdatert: 2018-01-11bibliografisk kontrollert
Holmgren, S., Nordén, M., Rantakokko, J. & Wallin, D. (2002). Performance of PDE solvers on a self-optimizing NUMA architecture. Parallel Algorithms and Applications, 17, 285-299
Åpne denne publikasjonen i ny fane eller vindu >>Performance of PDE solvers on a self-optimizing NUMA architecture
2002 (engelsk)Inngår i: Parallel Algorithms and Applications, ISSN 1063-7192, E-ISSN 1029-032X, Vol. 17, s. 285-299Artikkel i tidsskrift (Fagfellevurdert) Published
HSV kategori
Identifikatorer
urn:nbn:se:uu:diva-66909 (URN)10.1080/01495730208941445 (DOI)
Tilgjengelig fra: 2006-05-22 Laget: 2006-05-22 Sist oppdatert: 2018-01-10bibliografisk kontrollert
Organisasjoner