Many graph analysis methods are expressed via sparse matrix-matrix multiplication (SMMM). For example, the betweenness centrality, strong connected components, and minimum paths problems are solved through a series of sparse matrix-matrix multiplications. The adaptation of a particular SMMM algorithm to a specific application domain can be achieved by choosing an appropriate underlying semiring. In this talk we present a novel highly scalable SMMM algorithm based on a midpoint approach. We apply the algorithm over the real field semiring to the construction of the density matrix in electronic structure theory. The computational benefits of the algorithm come from the matrix distribution over processes and the communication pattern of the algorithm that follow the interaction structure of the atomic system. We use IBM Blue Gene/Q supercomputer to demonstrate the weak scalability of our SMMM algorithm using up to 50,653 MPI processes and strong scalability using up to 185,193 MPI processes. The talk focuses on the detail s of the algorithm published in http://dx.doi.org/10.1021/acs.jctc.5b00382.
The document describes sparse matrix reconstruction using a matrix completion algorithm. It begins with an overview of the matrix completion problem and formulation. It then describes the algorithm which uses soft-thresholding to impose a low-rank constraint and iteratively finds the matrix that agrees with the observed entries. The algorithm is proven to converge to the desired solution. Extensions to noisy data and generalized constraints are also discussed.
Math Geophysics-system of linear algebraic equationsAmin khalil
The document provides an overview of linear algebra concepts for mathematical geophysics, including:
- Definitions of equations, systems of linear algebraic equations, and the Gauss-Jordan reduction method.
- Types of systems include unique solution, no solution, and infinitely many solutions.
- Einstein summation convention simplifies tensor equations by implicitly summing over repeated indices.
- Gaussian elimination uses row operations to put a system of equations in row echelon form and then reduced row echelon form to solve for variables.
- Systems can have unique solutions, no solutions, or multiple solutions depending on the relationships between equations and variables.
A sparse matrix is a matrix with mostly zero elements, where a dense matrix has mostly non-zero elements. The sparsity of a sparse matrix is the percentage of zero elements it contains. Sparse matrices can save substantial memory by only storing the locations and values of non-zero elements, using formats like compressed row storage that use three one-dimensional arrays.
This document discusses solving sets of linear equations and analyzing DC circuits. It explains that to solve a set of linear equations in matrix form A*x=y, the matrix A and augmented matrix [A y] must have equal rank. It then discusses calculating the condition number of A, which should be close to 1 for an accurate solution. The document provides an example circuit in matrix form Z*I=V and shows using Matlab commands to calculate the rank of Z and [Z V], the condition number of Z, and solve for the current I.
1. This document discusses methods for solving linear algebraic equations and operations involving matrices. It covers topics such as matrix definitions, types of matrices, matrix operations, representing equations in matrix form, and methods for solving systems of linear equations including graphical methods, determinants, Cramer's rule, elimination, Gauss-Jordan, LU decomposition, and calculating the matrix inverse.
2. Key matrix operations include addition, multiplication, and rules for inverting a matrix. Methods for solving systems of equations include graphical techniques, determinants, Cramer's rule, elimination, Gauss, Gauss-Jordan, and LU decomposition.
3. LU decomposition involves writing a matrix as the product of a lower and upper triangular matrix, which can
System of linear algebriac equations nsmRahul Narang
The document discusses systems of linear algebraic equations and methods for solving them numerically. It introduces systems of linear equations in matrix form Ax = b and describes elementary row operations that can transform the matrix A. It then explains Gaussian elimination and Gauss-Jordan elimination methods for solving systems of linear equations by transforming the augmented matrix into reduced row echelon form. Finally, it briefly describes Jacobi and Gauss-Seidel iterative methods as well as applications of linear algebra in computer science fields like statistical learning, image manipulation, and physics.
The document describes sparse matrix reconstruction using a matrix completion algorithm. It begins with an overview of the matrix completion problem and formulation. It then describes the algorithm which uses soft-thresholding to impose a low-rank constraint and iteratively finds the matrix that agrees with the observed entries. The algorithm is proven to converge to the desired solution. Extensions to noisy data and generalized constraints are also discussed.
Math Geophysics-system of linear algebraic equationsAmin khalil
The document provides an overview of linear algebra concepts for mathematical geophysics, including:
- Definitions of equations, systems of linear algebraic equations, and the Gauss-Jordan reduction method.
- Types of systems include unique solution, no solution, and infinitely many solutions.
- Einstein summation convention simplifies tensor equations by implicitly summing over repeated indices.
- Gaussian elimination uses row operations to put a system of equations in row echelon form and then reduced row echelon form to solve for variables.
- Systems can have unique solutions, no solutions, or multiple solutions depending on the relationships between equations and variables.
A sparse matrix is a matrix with mostly zero elements, where a dense matrix has mostly non-zero elements. The sparsity of a sparse matrix is the percentage of zero elements it contains. Sparse matrices can save substantial memory by only storing the locations and values of non-zero elements, using formats like compressed row storage that use three one-dimensional arrays.
This document discusses solving sets of linear equations and analyzing DC circuits. It explains that to solve a set of linear equations in matrix form A*x=y, the matrix A and augmented matrix [A y] must have equal rank. It then discusses calculating the condition number of A, which should be close to 1 for an accurate solution. The document provides an example circuit in matrix form Z*I=V and shows using Matlab commands to calculate the rank of Z and [Z V], the condition number of Z, and solve for the current I.
1. This document discusses methods for solving linear algebraic equations and operations involving matrices. It covers topics such as matrix definitions, types of matrices, matrix operations, representing equations in matrix form, and methods for solving systems of linear equations including graphical methods, determinants, Cramer's rule, elimination, Gauss-Jordan, LU decomposition, and calculating the matrix inverse.
2. Key matrix operations include addition, multiplication, and rules for inverting a matrix. Methods for solving systems of equations include graphical techniques, determinants, Cramer's rule, elimination, Gauss, Gauss-Jordan, and LU decomposition.
3. LU decomposition involves writing a matrix as the product of a lower and upper triangular matrix, which can
System of linear algebriac equations nsmRahul Narang
The document discusses systems of linear algebraic equations and methods for solving them numerically. It introduces systems of linear equations in matrix form Ax = b and describes elementary row operations that can transform the matrix A. It then explains Gaussian elimination and Gauss-Jordan elimination methods for solving systems of linear equations by transforming the augmented matrix into reduced row echelon form. Finally, it briefly describes Jacobi and Gauss-Seidel iterative methods as well as applications of linear algebra in computer science fields like statistical learning, image manipulation, and physics.
IBM Graph – Graph Database-as-a-Service: Managing Data and Its Relationships ...Alexander Pozdneev
The slides presented by Alexander Pozdneev at GraphHPC-2017 conference (http://www.dislab.org/GraphHPC-2017/en/agenda.php).
Graph databases are increasingly popular in managing the information where the relationships between the data entities are of highest priority. However, a technical task of deploying, managing, and maintaining a graph database on-a-premise is decoupled from the process of solving an applied problem. IBM Graph is a graph database-as-a-service available on the IBM Bluemix cloud platform-as-a-service. IBM Graph is built upon open-source components while featuring high-availability and scalability on-demand. In this talk, we will introduce the main concepts behind IBM Graph and show how to leverage its API and the Bluemix console GUI.
Методология уточнения параметров системы разработки при планировании эксплуат...Alexander Pozdneev
В статье предлагается методология автоматизированного построения оптимальной (в смысле максимизации NPV) стратегии разработки нефтяного месторождения, учитывающая геологические неопределенности. На основе ансамбля гидродинамических моделей месторождения с использованием теории вероятностных графовых моделей строится стратегия, позволяющая в реальном времени принимать оптимальные решения о смене варианта сетки скважин, используя информацию, полученную от исследований на пилотных скважинах: пилотных стволах эксплуатационных скважин, а также скважинах опережающего бурения, которые бурятся с текущего куста на следующий для уточнения геологических целей. При этом в момент принятия решения не требуется проводить каких-либо дополнительных расчетов.
Предлагаемый подход был использован для построения оптимальной стратегии разработки одного из активов компании ПАО "Газпром нефть", находящегося на стадии эксплуатационного бурения. Разработка актива осложнена рядом неопределенностей, связанных со сложной неоднородной структурой коллектора и в высокой степени влияющих на выбор оптимальных технологических решений. Применение методологии позволило увеличить значение NPV на 15%.
A Methodology for the Refinement of Well Locations During Operational Drillin...Alexander Pozdneev
This document presents a methodology for refining well locations during operational drilling in the presence of geological uncertainty. The methodology involves:
1. Creating a strategy in advance that assesses potential development scenarios and defines rules for adapting the development system based on information from production drilling.
2. Generating a "technological tree" of possible development system solutions for a well pad that will undergo production drilling.
3. Testing the methodology on a real asset, where it increased expected NPV by 15% compared to the base case development system while maintaining the same recovery factor.
Enhanced MPSM3 for applications to quantum biological simulationsAlexander Pozdneev
This is the presentation for the talk that supported our SC16 paper (https://dl.acm.org/citation.cfm?id=3014916) - "Enhanced MPSM3 for Applications to Quantum Biological Simulations" by A. Pozdneev, V. Weber, T. Laino, C. Bekas, and A. Curioni, appeared in "Proceedings of SC16: The International Conference for High Performance Computing,
Networking, Storage and Analysis, Salt Lake City, Utah, November 13-18, 2016", Article no. 9.
Classical molecular dynamics simulations have been the preferred method to cope with the characteristic sizes and time scales of complex life-science systems. However, while classical methods have well known limitations, such as that their accuracy strongly depends on empirical tuning, the practical use of far more accurate methods that rely on quantum Hamiltonians, has been limited by the current efficiency and scalability of sparse matrix-matrix multiplication algorithms used in the self-consistent field equations. In this paper, we show unprecedented massive scalability of a recently presented method, called MPSM3, for sparse matrix-matrix multiplication. The algorithmic basis of the method was presented in a recent publication, while here we describe the algorithmic enhancements that allow us to claim at least one order of magnitude improvement in scalability and time to solution over the state of the art (original MPSM3). We achieve a time to solution for the multiplication of density matrices within the self-consistent field scheme that is approaching the time needed to evaluate energy and forces with classical force-field methods and that is independent from the system size, provided proportional resources. This latest development renders the application of entirely quantum Hamiltonians to systems of several millions of atoms for extended molecular dynamics investigations feasible.
We cover the IBM solution for HPC. In addition to hardware and software stack we show how the rational choice of compilation/running parameters helps to significantly improve the performance of technical computing applications.
Graph Community Detection Algorithm for Distributed Memory Parallel Computing...Alexander Pozdneev
Community detection is an important problem that spans many research areas, such as social networks, systems biology, power grid optimization. The fine-grained communication and irregular access pattern to memory and interconnect limit the overall scalability and performance of existing algorithms. This talk presents a highly scalable parallel algorithm for distributed memory systems. The method employs a novel implementation strategy to store and process dynamic graphs. The scalability analysis of the algorithm was performed using two massively parallel machines, Blue Gene/Q and Power7-IH, for graphs with up to hundreds of billions of edges. Leveraging the convergence properties of the algorithm and the efficient implementation, it is possible to analyze communities of large-scale graphs in just a few seconds. The talk is based on a paper accepted for publication in IPDPS-2015 conference proceedings that was kindly provided by Dr. Fabrizio Petrini (IBM Research).
Параллельные алгоритмы IBM Research для решения задач обхода и построения кра...Alexander Pozdneev
В докладе будет дан обзор второго поколения алгоритмов BFS, с помощью которых были получены результаты для систем IBM Blue Gene/Q Mira и Sequoia, отраженные в редакциях рейтинга Graph500 за 2012-2013 годы. Будут описаны структура для хранения графа, схема декомпозиции данных, принципы балансировки нагрузки и методы, позволяющие в ряде случаев на порядок сократить число ребер, которые необходимо рассмотреть в процессе обхода. Вторая часть доклада будет посвящена новому параллельному алгоритму решения задачи SSSP. Алгоритм характеризуется существенно меньшим по сравнению с известными алгоритмами объемом межпроцессорных коммуникаций. Будут рассмотрены классы оптимизаций, направленные на решение проблемы балансировки нагрузки для больших графов, а также вопросы уменьшения количества итераций алгоритма и числа релаксаций ребер. В докладе будет показано влияние отдельных приемов оптимизации на производительность алгоритмов BFS и SSSP на системах архитектуры IBM Blue Gene/Q. Материалы для доклада любезно предоставлены доктором Fabrizio Petrini (IBM Research) и будут опубликованы в трудах конференции IPDPS-2014.
Литература
* Traversing Trillions of Edges in Real-time: Graph Exploration on Large-scale Parallel Machines
Fabio Checconi, Fabrizio Petrini
http://www.odbms.org/wp-content/uploads/2014/05/g500-ipdps14.pdf
* Scalable Single Source Shortest Path Algorithms for Massively Parallel Systems
Venkatesan T. Chakaravarthy, Fabio Checconi, Fabrizio Petrini, Yogish Sabharwal
http://www.odbms.org/wp-content/uploads/2014/05/sssp-ipdps2014.pdf
More Related Content
Similar to Midpoint-Based Parallel Sparse Matrix-Matrix Multiplication Algorithm
IBM Graph – Graph Database-as-a-Service: Managing Data and Its Relationships ...Alexander Pozdneev
The slides presented by Alexander Pozdneev at GraphHPC-2017 conference (http://www.dislab.org/GraphHPC-2017/en/agenda.php).
Graph databases are increasingly popular in managing the information where the relationships between the data entities are of highest priority. However, a technical task of deploying, managing, and maintaining a graph database on-a-premise is decoupled from the process of solving an applied problem. IBM Graph is a graph database-as-a-service available on the IBM Bluemix cloud platform-as-a-service. IBM Graph is built upon open-source components while featuring high-availability and scalability on-demand. In this talk, we will introduce the main concepts behind IBM Graph and show how to leverage its API and the Bluemix console GUI.
Методология уточнения параметров системы разработки при планировании эксплуат...Alexander Pozdneev
В статье предлагается методология автоматизированного построения оптимальной (в смысле максимизации NPV) стратегии разработки нефтяного месторождения, учитывающая геологические неопределенности. На основе ансамбля гидродинамических моделей месторождения с использованием теории вероятностных графовых моделей строится стратегия, позволяющая в реальном времени принимать оптимальные решения о смене варианта сетки скважин, используя информацию, полученную от исследований на пилотных скважинах: пилотных стволах эксплуатационных скважин, а также скважинах опережающего бурения, которые бурятся с текущего куста на следующий для уточнения геологических целей. При этом в момент принятия решения не требуется проводить каких-либо дополнительных расчетов.
Предлагаемый подход был использован для построения оптимальной стратегии разработки одного из активов компании ПАО "Газпром нефть", находящегося на стадии эксплуатационного бурения. Разработка актива осложнена рядом неопределенностей, связанных со сложной неоднородной структурой коллектора и в высокой степени влияющих на выбор оптимальных технологических решений. Применение методологии позволило увеличить значение NPV на 15%.
A Methodology for the Refinement of Well Locations During Operational Drillin...Alexander Pozdneev
This document presents a methodology for refining well locations during operational drilling in the presence of geological uncertainty. The methodology involves:
1. Creating a strategy in advance that assesses potential development scenarios and defines rules for adapting the development system based on information from production drilling.
2. Generating a "technological tree" of possible development system solutions for a well pad that will undergo production drilling.
3. Testing the methodology on a real asset, where it increased expected NPV by 15% compared to the base case development system while maintaining the same recovery factor.
Enhanced MPSM3 for applications to quantum biological simulationsAlexander Pozdneev
This is the presentation for the talk that supported our SC16 paper (https://dl.acm.org/citation.cfm?id=3014916) - "Enhanced MPSM3 for Applications to Quantum Biological Simulations" by A. Pozdneev, V. Weber, T. Laino, C. Bekas, and A. Curioni, appeared in "Proceedings of SC16: The International Conference for High Performance Computing,
Networking, Storage and Analysis, Salt Lake City, Utah, November 13-18, 2016", Article no. 9.
Classical molecular dynamics simulations have been the preferred method to cope with the characteristic sizes and time scales of complex life-science systems. However, while classical methods have well known limitations, such as that their accuracy strongly depends on empirical tuning, the practical use of far more accurate methods that rely on quantum Hamiltonians, has been limited by the current efficiency and scalability of sparse matrix-matrix multiplication algorithms used in the self-consistent field equations. In this paper, we show unprecedented massive scalability of a recently presented method, called MPSM3, for sparse matrix-matrix multiplication. The algorithmic basis of the method was presented in a recent publication, while here we describe the algorithmic enhancements that allow us to claim at least one order of magnitude improvement in scalability and time to solution over the state of the art (original MPSM3). We achieve a time to solution for the multiplication of density matrices within the self-consistent field scheme that is approaching the time needed to evaluate energy and forces with classical force-field methods and that is independent from the system size, provided proportional resources. This latest development renders the application of entirely quantum Hamiltonians to systems of several millions of atoms for extended molecular dynamics investigations feasible.
We cover the IBM solution for HPC. In addition to hardware and software stack we show how the rational choice of compilation/running parameters helps to significantly improve the performance of technical computing applications.
Graph Community Detection Algorithm for Distributed Memory Parallel Computing...Alexander Pozdneev
Community detection is an important problem that spans many research areas, such as social networks, systems biology, power grid optimization. The fine-grained communication and irregular access pattern to memory and interconnect limit the overall scalability and performance of existing algorithms. This talk presents a highly scalable parallel algorithm for distributed memory systems. The method employs a novel implementation strategy to store and process dynamic graphs. The scalability analysis of the algorithm was performed using two massively parallel machines, Blue Gene/Q and Power7-IH, for graphs with up to hundreds of billions of edges. Leveraging the convergence properties of the algorithm and the efficient implementation, it is possible to analyze communities of large-scale graphs in just a few seconds. The talk is based on a paper accepted for publication in IPDPS-2015 conference proceedings that was kindly provided by Dr. Fabrizio Petrini (IBM Research).
Параллельные алгоритмы IBM Research для решения задач обхода и построения кра...Alexander Pozdneev
В докладе будет дан обзор второго поколения алгоритмов BFS, с помощью которых были получены результаты для систем IBM Blue Gene/Q Mira и Sequoia, отраженные в редакциях рейтинга Graph500 за 2012-2013 годы. Будут описаны структура для хранения графа, схема декомпозиции данных, принципы балансировки нагрузки и методы, позволяющие в ряде случаев на порядок сократить число ребер, которые необходимо рассмотреть в процессе обхода. Вторая часть доклада будет посвящена новому параллельному алгоритму решения задачи SSSP. Алгоритм характеризуется существенно меньшим по сравнению с известными алгоритмами объемом межпроцессорных коммуникаций. Будут рассмотрены классы оптимизаций, направленные на решение проблемы балансировки нагрузки для больших графов, а также вопросы уменьшения количества итераций алгоритма и числа релаксаций ребер. В докладе будет показано влияние отдельных приемов оптимизации на производительность алгоритмов BFS и SSSP на системах архитектуры IBM Blue Gene/Q. Материалы для доклада любезно предоставлены доктором Fabrizio Petrini (IBM Research) и будут опубликованы в трудах конференции IPDPS-2014.
Литература
* Traversing Trillions of Edges in Real-time: Graph Exploration on Large-scale Parallel Machines
Fabio Checconi, Fabrizio Petrini
http://www.odbms.org/wp-content/uploads/2014/05/g500-ipdps14.pdf
* Scalable Single Source Shortest Path Algorithms for Massively Parallel Systems
Venkatesan T. Chakaravarthy, Fabio Checconi, Fabrizio Petrini, Yogish Sabharwal
http://www.odbms.org/wp-content/uploads/2014/05/sssp-ipdps2014.pdf
1. Midpoint-Based Parallel Sparse Matrix–Matrix
Multiplication Algorithm
V. Weber1, A. Pozdneev2, T. Laino1
1
IBM Zurich Research Laboratory, R¨uschlikon, Switzerland
2
IBM Science and Technology Center, Moscow, Russia
March 3, 2016 — GraphHPC-2016, Moscow
1 c○ 2016 IBM Corporation
2. Matrix graph duality
Graph algorithms in the language of linear algebra
∙ Adjacency matrix A is dual with the corresponding graph
∙ Vector matrix multiply is dual with breadth-first search
Image credit: [Kepner2011]
2 c○ 2016 IBM Corporation
3. Sparse matrix-matrix multiplication motivation
Electronic structure theory
∙ Density matrix 𝐷 at zero
electronic temperature:
𝐷 = 𝜃[𝜇𝐼 − 𝐹]
𝜃(𝑥) — Heaviside step
function
𝜇 — chemical potential
𝐼 — identity matrix
𝐹 — Fockian
∙ Second-order spectral projection (SP2)
method
𝐷 = lim
𝑖→∞
𝑓𝑖[𝑓𝑖−1[. . . 𝑓0[𝑋0]]]
∙ Recursive polynomial
𝑓𝑖[𝑋𝑖] =
{︃
𝑋2
𝑖 Tr[𝑋𝑖] > 𝑁 𝑠
2𝑋𝑖 − 𝑋2
𝑖 otherwise
𝑁 𝑠 — number of occupied states
3 c○ 2016 IBM Corporation
4. Scalability issues of the Cannon’s and SUMMA algorithms
∙ Weak scaling experiment
∙ ≈ 200 atoms per MPI task
∙ From 32 928 atoms on 144 cores
∙ To 1 022 208 atoms on 5184 cores
∙ Image legend:
Total CPU time per atom
Block matrix multiplications
Book-keeping overhead
Communication between processes
Dashed line: fit to the data using
𝑓(𝑁) = 𝑎 + 𝑏
√
𝑁
Image credit: [VandeVondele2012]
4 c○ 2016 IBM Corporation
5. Midpoint method
∙ Particle interactions in molecular
dynamics [Bowers2006]
∙ The computational domain is
divided between processors
∙ Interaction between a pair — a
processor that holds a midpoint
Image credit: [Bowers2006]
5 c○ 2016 IBM Corporation
8. Sparse matrix-matrix multiplication
∙ Exploit the locality
∙ 3D Cartesian topology
∙ Each matrix element lies at the
process that owns a midpoint:
midpoint(𝑖, 𝑘) ⇒ element 𝐴𝑖𝑘
midpoint(𝑘, 𝑗) ⇒ element 𝐵 𝑘𝑗
midpoint(𝑖, 𝑗) ⇒ element 𝐶𝑖𝑗
∙ Send 𝐴𝑖𝑘 and 𝐵 𝑘𝑗
∙ 𝐶𝑖𝑗 ← 𝐴𝑖𝑘 · 𝐵 𝑘𝑗 + 𝐶𝑖𝑗
i
j
k
Bkj
(-1,0) (0,0)
(0,-1)
Cij
Aik
Image credit: [Weber2015]
8 c○ 2016 IBM Corporation
9. Why does it work?
As a reminder. . .
Midpoint distribution of a matrix
𝐴 = 𝐴(1,1)
+ · · · + 𝐴(𝑝 𝑥,𝑝 𝑦)
𝐴𝐵 = (𝐴(1,1)
+ · · · + 𝐴(𝑝 𝑥,𝑝 𝑦)
) ×
× (𝐵(1,1)
+ · · · + 𝐵(𝑝 𝑥,𝑝 𝑦)
)
𝐴(𝑟,𝑠)
𝐵(𝑡,𝑢)
̸= 𝑂
if |𝑟 − 𝑡| ≤ 2, |𝑠 − 𝑢| ≤ 2
9 c○ 2016 IBM Corporation
10. Weak scaling experiments
∙ Two occupations of the
density matrix (O1, O2)
∙ ≈ 19 water molecules
per MPI task
∙ 216 ÷ 59 319 MPI tasks
∙ MPSM3 vs. SUMMA
crossover point:
O1: ≈ 19 683 MPI
tasks (373 248 water
molecules)
O2: ≈ 9 261 MPI
tasks (175 616 water
molecules)
Image credit: [Weber2015]
10 c○ 2016 IBM Corporation
11. Strong scaling experiments
∙ Water molecules:
S1: 110 592
S2: 373 248
S3: 1 124 864
∙ MPI tasks:
S1: 216 ÷ 59 319
S2: 1 728 ÷ 110 592
S3: 9 261 ÷ 185 193
∙ MPSM3 vs. SUMMA
crossover point — slightly
smaller than S2 Image credit: [Weber2015]
11 c○ 2016 IBM Corporation
12. Summary
∙ Linear algebra is the language of graph algorithms
∙ Sparse matrix-matrix multiplication is a critical building block
∙ General-purpose SMMM algorithms do not scale well
∙ We distribute sparse matrix according to the midpoint principle
∙ Linear scalability with number of “edges” given proportional resources
∙ Advantages over SUMMA:
reduced communication volume
more effective load balancing
communication latency (nearby processes only)
∙ The scalability comes from the reduced volume of interprocess
communications
12 c○ 2016 IBM Corporation
13. References
V. Weber, T. Laino, A. Pozdneev, I. Fedulova, A. Curioni
Semiempirical Molecular Dynamics (SEMD) I: Midpoint-Based Parallel
Sparse Matrix–Matrix Multiplication Algorithm for Matrices with Decay
J. Chem. Theory Comput., 2015, 11 (7), pp 3145–3152
K.J. Bowers, R.O. Dror, D.E. Shaw
The midpoint method for parallelization of particle simulations
J. Chem. Phys., 124, 184109 (2006)
J. VandeVondele, U. Borˇstnik, J. Hutter
Linear Scaling Self-Consistent Field Calculations with Millions of Atoms
in the Condensed Phase
J. Chem. Theory Comput., 2012, 8 (10), pp 3565–3573
J. Kepner, J. Gilbert (eds.)
Graph Algorithms in the Language of Linear Algebra
SIAM, Philadelphia, 2011
13 c○ 2016 IBM Corporation
14. Disclaimer
All the information, representations, statements, opinions and proposals in this
document are correct and accurate to the best of our present knowledge but are
not intended (and should not be taken) to be contractually binding unless and
until they become the subject of separate, specific agreement between us.
Any IBM Machines provided are subject to the Statements of Limited Warranty
accompanying the applicable Machine.
Any IBM Program Products provided are subject to their applicable license terms.
Nothing herein, in whole or in part, shall be deemed to constitute a warranty.
IBM products are subject to withdrawal from marketing and or service upon
notice, and changes to product configurations, or follow-on products, may result
in price changes.
Any references in this document to “partner” or “partnership” do not constitute or
imply a partnership in the sense of the Partnership Act 1890.
IBM is not responsible for printing errors in this proposal that result in pricing or
information inaccuracies.
14 c○ 2016 IBM Corporation
15. Правовая информация
IBM, логотип IBM, BladeCenter, System Storage и System x являются товарными знаками International Business
Machines Corporation в США и/или других странах. Полный список товарных знаков компании IBM смотрите
на узле Web: www.ibm.com/legal/copytrade.shtml.
Названия других компаний, продуктов и услуг могут являться товарными знаками или знаками обслуживания
других компаний.
(c) 2016 International Business Machines Corporation. Все права защищены.
Упоминание в этой публикации продуктов или услуг корпорации IBM не означает, что IBM предполагает
предоставлять их во всех странах, в которых осуществляет свою деятельность, информация о
предоставлении продуктов или услуг может быть изменена без уведомления. За самой свежей информацией
о продуктах и услугах компании IBM, предоставляемых в Вашем регионе, следует обращаться в ближайшее
торговое представительство IBM или к авторизованным бизнес-партнерам.
Все заявления относительно намерений и перспективных планов IBM могут быть изменены без уведомления.
Информация о продуктах третьих фирм получена от производителей этих продуктов или из опубликованных
анонсов указанных продуктов. IBM не тестировала эти продукты и не может подтвердить
производительность, совместимость, или любые другие заявления относительно продуктов третьих фирм.
Вопросы о возможностях продуктов третьих фирм следует адресовать поставщику этих продуктов.
Информация может содержать технические неточности или типографические ошибки. В представленную в
публикации информацию могут вноситься изменения, эти изменения будут включаться в новые редакции
данной публикации. IBM может вносить изменения в рассматриваемые в данной публикации продукты или
услуги в любое время без уведомления.
Любые ссылки на узлы Web третьих фирм приведены только для удобства и никоим образом не служат
поддержкой этим узлам Web. Материалы на указанных узлах Web не являются частью материалов для
данного продукта IBM.
15 c○ 2016 IBM Corporation