umu.sePublikasjoner
Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Identification and tuning of algorithmic parameters in parallel matrix computations: Hessenberg reduction and tensor storage format conversion
Umeå universitet, Teknisk-naturvetenskapliga fakulteten, Institutionen för datavetenskap.
2018 (engelsk)Licentiatavhandling, med artikler (Annet vitenskapelig)
Abstract [en]

This thesis considers two problems in numerical linear algebra and high performance computing (HPC): (i) the parallelization of a new blocked Hessenberg reduction algorithm using Parallel Cache Assignment (PCA) and the tunability of its algorithm parameters, and (ii) storing and manipulating dense tensors on shared memory HPC systems.

The Hessenberg reduction appears in the Aggressive Early Deflation (AED) process for identifying converged eigenvalues in the distributed multishift QR algorithm (state-of-the-art algorithm for computing all eigenvalues for dense square matrices). Since the AED process becomes a parallel bottleneck it motivates a further study of AED components. We present a new Hessenberg reduction algorithm based on PCA which is NUMA-aware and targeting relatively small problem sizes on shared memory systems. The tunability of the algorithm parameters are investigated. A simple off-line tuning is presented and the performance of the new Hessenberg reduction algorithm is compared to its counterparts from LAPACK and ScaLAPACK. The new algorithm outperforms LAPACK in all tested cases and outperforms ScaLAPACK in problems smaller than order 1500, which are common problem sizes for AED in the context of the distributed multishift QR algorithm.

We also investigate automatic tuning of the algorithm parameters. The parameters span a huge search space and it is impractical to tune them using standard auto-tuning and optimization techniques. We present a modular auto-tuning framework which applies: search space decomposition, binning, and multi-stage search to enable searching the huge search space efficiently. The framework using these techniques exposes the underlying subproblems which allows using standard auto-tuning methods to tune them. In addition, the framework defines an abstract interface, which combined with its modular design, allows testing various tuning algorithms.

In the last part of the thesis, the focus is on the problem of storing and manipulating dense tensors. Developing open source tensor algorithms and applications is hard due to the lack of open source software for fundamental tensor operations. We present a software library dten, which includes tools for storing dense tensors in shared memory and converting a tensor storage format from one canonical form to another. The library provides two different ways to perform the conversion in parallel, in-place and out-of-place. The conversion involves moving blocks of contiguous data and are done to maximize the size of the blocks to move. In addition, the library supports tensor matricization for one or two tensors at the same time. The latter case is important in preparing tensors for contraction operations. The library is general purpose and highly flexible.

sted, utgiver, år, opplag, sider
Umeå: Umeå universitet , 2018. , s. 15
Serie
Report / UMINF, ISSN 0348-0542 ; UMINF 18.22
HSV kategori
Identifikatorer
URN: urn:nbn:se:umu:diva-145345ISBN: 978-91-7601-843-9 (tryckt)OAI: oai:DiVA.org:umu-145345DiVA, id: diva2:1186623
Veileder
Tilgjengelig fra: 2018-03-01 Laget: 2018-02-28 Sist oppdatert: 2018-06-09bibliografisk kontrollert
Delarbeid
1. On the Tunability of a New Hessenberg Reduction Algorithm Using Parallel Cache Assignment
Åpne denne publikasjonen i ny fane eller vindu >>On the Tunability of a New Hessenberg Reduction Algorithm Using Parallel Cache Assignment
2018 (engelsk)Inngår i: Parallel Processing and Applied Mathematics. PPAM 2017: Part 1 / [ed] Wyrzykowski R., Dongarra J., Deelman E., Karczewski K., Springer, 2018, s. 579-589Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

The reduction of a general dense square matrix to Hessenberg form is a well known first step in many standard eigenvalue solvers. Although parallel algorithms exist, the Hessenberg reduction is one of the bottlenecks in AED, a main part in state-of-the-art software for the distributed multishift QR algorithm. We propose a new NUMA-aware algorithm that fits the context of the QR algorithm and evaluate the sensitivity of its algorithmic parameters. The proposed algorithm is faster than LAPACK for all problem sizes and faster than ScaLAPACK for the relatively small problem sizes typical for AED.

sted, utgiver, år, opplag, sider
Springer, 2018
Serie
Lecture Notes in Computer Science, ISSN 0302-9743, E-ISSN 1611-3349 ; 10777
Emneord
Hessenberg reduction, Parallel cache assignment, NUMA-aware algorithm, Shared-memory, Tunable parameters, Off-line tuning
HSV kategori
Identifikatorer
urn:nbn:se:umu:diva-145342 (URN)10.1007/978-3-319-78024-5_50 (DOI)000458563300050 ()978-3-319-78023-8 (ISBN)978-3-319-78024-5 (ISBN)
Konferanse
12th International Conference on Parallel Processing and Applied Mathematics, PPAM 2017, Lublin, Poland, 10–13 September, 2017
Merknad

Tilgjengelig fra: 2018-02-28 Laget: 2018-02-28 Sist oppdatert: 2019-03-06bibliografisk kontrollert
2. An auto-tuning framework for a NUMA-aware Hessenberg reduction algorithm
Åpne denne publikasjonen i ny fane eller vindu >>An auto-tuning framework for a NUMA-aware Hessenberg reduction algorithm
2017 (engelsk)Rapport (Annet vitenskapelig)
Abstract [en]

The performance of a recently developed Hessenberg reduction algorithm greatly depends on the values chosen for its tunable parameters. The search space is huge combined with other complications makes the problem hard to solve effectively with generic methods and tools. We describe a modular auto-tuning framework in which the underlying optimization algorithm is easy to substitute. The framework exposes sub-problems of standard auto-tuning type for which existing generic methods can be reused. The outputs of concurrently executing sub-tuners are assembled by the framework into a solution to the original problem.

sted, utgiver, år, opplag, sider
Umeå: Department of computing science, Umeå university, 2017. s. 14
Serie
Report / UMINF, ISSN 0348-0542 ; 17.19
Emneord
Auto-tuning, Tuning framework, Binning, Search space decomposition, Multistage search, Hessenberg reduction, NUMA-aware
HSV kategori
Identifikatorer
urn:nbn:se:umu:diva-145297 (URN)
Tilgjengelig fra: 2018-02-28 Laget: 2018-02-28 Sist oppdatert: 2018-06-09bibliografisk kontrollert
3. A library for storing and manipulating dense tensors
Åpne denne publikasjonen i ny fane eller vindu >>A library for storing and manipulating dense tensors
2016 (engelsk)Rapport (Annet vitenskapelig)
Abstract [en]

Aiming to build a layered infrastructure for high-performance dense tensor applications, we present a library, called dten, for storing and manipulating dense tensors. The library focuses on storing dense tensors in canonical storage formats and converting between storage formats in parallel. In addition, it supports tensor matricization in different ways. The library is general-purpose and provides a high degree of flexibility.

sted, utgiver, år, opplag, sider
Umeå: Department of computing science, Umeå university, 2016. s. 21
Serie
Report / UMINF, ISSN 0348-0542 ; 17.22
Emneord
Dense tensors, canonical storage format, tensor matricization, tensor storage format conversion, out-of-place conversion, in-place conversion
HSV kategori
Identifikatorer
urn:nbn:se:umu:diva-145341 (URN)
Tilgjengelig fra: 2018-02-28 Laget: 2018-02-28 Sist oppdatert: 2018-06-09bibliografisk kontrollert

Open Access i DiVA

fulltext(149 kB)48 nedlastinger
Filinformasjon
Fil FULLTEXT01.pdfFilstørrelse 149 kBChecksum SHA-512
babcbd738e503c09f2ae79ed781b9fb6ed8a4fc85a9956097a3c68465312f67bf69afacc73aacfd9b491981c1cb70ec0636d52bb5b3ea83d63a3edb16bb84aab
Type fulltextMimetype application/pdf

Personposter BETA

Eljammaly, Mahmoud

Søk i DiVA

Av forfatter/redaktør
Eljammaly, Mahmoud
Av organisasjonen

Søk utenfor DiVA

GoogleGoogle Scholar
Totalt: 48 nedlastinger
Antall nedlastinger er summen av alle nedlastinger av alle fulltekster. Det kan for eksempel være tidligere versjoner som er ikke lenger tilgjengelige

isbn
urn-nbn

Altmetric

isbn
urn-nbn
Totalt: 500 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf