Bridging cultures that have often been distant, Julia combines expertise from the diverse fields of computer science and computational science to create a new approach to numerical computing. Julia is designed to be easy and fast and questions notions generally held to be “laws of nature" by practitioners of numerical computing: \beginlist \item High-level dynamic programs have to be slow. \item One must prototype in one language and then rewrite in another language for speed or deployment. \item There are parts of a system appropriate for the programmer, and other parts that are best left untouched as they have been built by the experts. \endlist We introduce the Julia programming language and its design---a dance between specialization and abstraction. Specialization allows for custom treatment. Multiple dispatch, a technique from computer science, picks the right algorithm for the right circumstance. Abstraction, which is what good computation is really about, recognizes what remains the same after dif...

Julia: A Fresh Approach to Numerical Computing

Bridging cultures that have often been distant, Julia combines expertise from the diverse fields of computer science and computational science to create a new approach to numerical computing. Julia is designed to be easy and fast. Julia questions notions generally held as "laws of nature" by practitioners of numerical computing: 
1. High-level dynamic programs have to be slow. 
2. One must prototype in one language and then rewrite in another language for speed or deployment, and 
3. There are parts of a system for the programmer, and other parts best left untouched as they are built by the experts. 
We introduce the Julia programming language and its design --- a dance between specialization and abstraction. Specialization allows for custom treatment. Multiple dispatch, a technique from computer science, picks the right algorithm for the right circumstance. Abstraction, what good computation is really about, recognizes what remains the same after differences are stripped away. Abstractions in mathematics are captured as code through another technique from computer science, generic programming. 
Julia shows that one can have machine performance without sacrificing human convenience.

Training of large-scale deep neural networks is often constrained by the available computational resources. We study the effect of limited precision data representation and computation on neural network training. Within the context of low-precision fixed-point computations, we observe the rounding scheme to play a crucial role in determining the network's behavior during training. Our results show that deep networks can be trained using only 16-bit wide fixed-point number representation when using stochastic rounding, and incur little to no degradation in the classification accuracy. We also demonstrate an energy-efficient hardware accelerator that implements low-precision fixed-point arithmetic with stochastic rounding.

Deep Learning with Limited Numerical Precision

Training of large-scale deep neural networks is often constrained by the available computational resources. We study the effect of limited precision data representation and computation on neural network training. Within the context of lowprecision fixed-point computations, we observe the rounding scheme to play a crucial role in determining the network's behavior during training. Our results show that deep networks can be trained using only 16-bit wide fixed-point number representation when using stochastic rounding, and incur little to no degradation in the classification accuracy. We also demonstrate an energy-efficient hardware accelerator that implements low-precision fixed-point arithmetic with stochastic rounding.

/pdf/deep-learning-with-limited-numerical-precision-5e695rwxnp.pdf

As multicore systems continue to gain ground in the high performance computing world, linear algebra algorithms have to be reformulated or new algorithms have to be developed in order to take advantage of the architectural features on these new processors. Fine grain parallelism becomes a major requirement and introduces the necessity of loose synchronization in the parallel execution of an operation. This paper presents algorithms for the Cholesky, LU and QR factorization where the operations can be represented as a sequence of small tasks that operate on square blocks of data. These tasks can be dynamically scheduled for execution based on the dependencies among them and on the availability of computational resources. This may result in out of order execution of tasks which will completely hide the presence of intrinsically sequential tasks in the factorization. Performance comparisons are presented with LAPACK algorithms where parallelism can only be exploited at the level of the BLAS operations and vendor implementations.

/pdf/a-class-of-parallel-tiled-linear-algebra-algorithms-for-362zg19tsx.pdf

A class of parallel tiled linear algebra algorithms for multicore architectures

The emergence and continuing use of multi-core architectures and graphics processing units require changes in the existing software and sometimes even a redesign of the established algorithms in order to take advantage of now prevailing parallelism. Parallel Linear Algebra for Scalable Multi-core Architectures (PLASMA) and Matrix Algebra on GPU and Multics Architectures (MAGMA) are two projects that aims to achieve high performance and portability across a wide range of multi-core architectures and hybrid systems respectively. We present in this document a comparative study of PLASMA's performance against established linear algebra packages and some preliminary results of MAGMA on hybrid multi-core and GPU systems.

http://icl.cs.utk.edu/news_pub/submissions/plasma-scidac09.pdf

Numerical linear algebra on emerging architectures: The PLASMA and MAGMA projects

We present parallel and sequential dense QR factorization algorithms that are both optimal (up to polylogarithmic factors) in the amount of communication they perform and just as stable as Householder QR. We prove optimality by deriving new lower bounds for the number of multiplications done by “non-Strassen-like” QR, and using these in known communication lower bounds that are proportional to the number of multiplications. We not only show that our QR algorithms attain these lower bounds (up to polylogarithmic factors), but that existing LAPACK and ScaLAPACK algorithms perform asymptotically more communication. We derive analogous communication lower bounds for LU factorization and point out recent LU algorithms in the literature that attain at least some of these lower bounds. The sequential and parallel QR algorithms for tall and skinny matrices lead to significant speedups in practice over some of the existing algorithms, including LAPACK and ScaLAPACK, for example, up to 6.7 times over ScaLAPACK. A performance model for the parallel algorithm for general rectangular matrices predicts significant speedups over ScaLAPACK.

/pdf/communication-optimal-parallel-and-sequential-qr-and-lu-1ldw6a0wk4.pdf

Communication-optimal Parallel and Sequential QR and LU Factorizations

We present parallel and sequential dense QR factorization algorithms that are both optimal (up to polylogarithmic factors) in the amount of communication they perform, and just as stable as Householder QR. 
We prove optimality by extending known lower bounds on communication bandwidth for sequential and parallel matrix multiplication to provide latency lower bounds, and show these bounds apply to the LU and QR decompositions. We not only show that our QR algorithms attain these lower bounds (up to polylogarithmic factors), but that existing LAPACK and ScaLAPACK algorithms perform asymptotically more communication. We also point out recent LU algorithms in the literature that attain at least some of these lower bounds.

Communication-optimal parallel and sequential QR and LU factorizations

http://www.netlib.org/utk/people/JackDongarra/PAPERS/bddl08.pdf

Julien Langou

Papers

A class of parallel tiled linear algebra algorithms for multicore architectures

Numerical linear algebra on emerging architectures: The PLASMA and MAGMA projects

Communication-optimal Parallel and Sequential QR and LU Factorizations

Communication-optimal parallel and sequential QR and LU factorizations

Algorithm-based fault tolerance applied to high performance computing