Questions on parallel computing in Mathematica.

learn more… | top users | synonyms

9
votes
0answers
258 views

Using Compile to speed up Function with PermutationProduct

I have the following fairly simple routine dot involving PermutationProduct that I wish to speed up using ...
7
votes
0answers
217 views

Partial Differential Equation in Parallel

is there any native way to implement multi-core parallel solving of PDE in Wolfram Mathematica? WM 10 now supports Finite Elements Method, but it is actually useless without parallelization. Usually ...
5
votes
0answers
56 views

The change of precision doesn't work in ParallelDo. Is it a bug in Parallel?

Cross-post: http://zhidao.baidu.com/question/2075301703219376588.html Input: ...
5
votes
0answers
1k views

Threads vs Cores: How many kernels need to launch in parallel Mathematica?

I have a dual-core intel machine. I would like to test how many parallel local kernels should be launch to get the optimized performance. Knowing that the HyperThreading technology of Intel could run ...
5
votes
0answers
122 views

What do the elements of the Parallel Kernel Status screen mean?

I just managed to get my multiple kernels up and humming on a new multi-core machine, which I am quite happy about. I am slightly at a loss, though, about the meaning of the different parts of the ...
5
votes
0answers
264 views

Parallelizing is much slower than single kernel evaluation?

Starting with a vector of matrices, tVec, I would like to parallelize the multiplication of each element with the same matrix ...
5
votes
0answers
164 views

Raster with ColorFunction is blank when Dynamic Updating is disabled

UPDATE Today I was having these WaveletScalogram problems again. I think this is ultimately some kind of interaction issue with Dynamic. I have narrowed it down to ...
4
votes
0answers
226 views

Mathematica + RasPi + Parallela = Kaboom?

Is it possible to use the Parallela computing platform to enhance Mathematica computing on a Raspberry Pi? It's not at all clear to me how to interface the two boards (here it says it's done via ...
4
votes
0answers
217 views

OpenCL vs. CUDA with Mathematica

I am thinking about using GPU programming with Mathematica using a MacBook Pro with a NVIDIA GPU. At this page, the Mathematica documentation says "Programming OpenCL in the Wolfram Language is simple ...
4
votes
0answers
210 views

How to force the compilation language to be C++ (not C) in CreateExecutable or CreateLibrary

I'm trying to speed up a section of a Mathematica 9.0.1 notebook by performing some calculations in C++ with the Cilk Plus and Threading Building Blocks tools available for the Intel C++ 14.0 compiler ...
4
votes
0answers
197 views

Discriminant of Characteristic Polynomial

I'm doing a calculation which finds the characteristic polynomial of a matrix with rather complex entries and then determines the discriminant of that polynomial. For smaller matrices up to around 7x7 ...
4
votes
0answers
113 views

How can you link Parallel Mapping and Error tracking

Following this question and the comments made, together with the information in this question, the following code allows error tracking and debugging very nicely. ...
3
votes
0answers
55 views

AppendTo or Part in a ParallelDo?

I have a question regarding how memory is managed in parallel. Specifically, I'd like to know what is the better practice when constructing an array within a ...
3
votes
0answers
56 views

Smart use of ParallelMap over Plus

I'm looking for ways to speed up calculation of a integral with a many-terms integrant. The obvious way is to go in parallel. I was surprised, but for some tests integration by ...
3
votes
0answers
76 views

k-means clustering in parallel

How can the k-means algorithm for some given distance function be implemented using ParallelMap or other parallel processing program constructs in Mathematica?
3
votes
0answers
90 views

LibraryLink parallel performance

I'm trying to use a LibraryLink function in parallel. It is an embarrassingly parallel problem, so that no dependency between those parallel tasks. However I can only get the parallel efficiency about ...
3
votes
0answers
179 views

Launching Remote Kernels takes a lot of time

I recently played around with establishing a remote kernel setup without using mathematica grid. To do this I use the LaunchKernels[] command and ...
3
votes
0answers
307 views

To what extent can NonlinearModelFit be sped up?

Can NonlinearModelFit be parallelized or compiled to allow one to achieve a scalable speedup for a fitting procedure? Attempting to use this function to fit a ...
3
votes
0answers
186 views

Mathlink and multithreading

In order to make my computations faster I have developed a C/C++ application calling Mathematica kernel and performing many operations using Mathematica functions for equation solving. Now I want to ...
2
votes
0answers
40 views

Parallel evaluation of symbolic matrix eigenvalues

I'm trying to find eigenvalues of symbolic hermitian 40x40 block matrix. It is made from 20 2x2 matrices like this: ArrayFlatten[Table[M[m,n],{m,1,20},{n,1,20}]] ...
2
votes
0answers
54 views

Is it possible to automatically apply Parallelize when possible

Parallelize in Mathematica is pretty easy to use. Most of the time, applying Parallelize to an expression, it tries to evaluates ...
2
votes
0answers
41 views

Export on local machine results of a parallel computation using remote kernels

I've got a long calculus to do with Mathematica. I've Parallelized the code using local kernels and remote kernels. I'm using a ParallelDo loop with an ...
2
votes
0answers
28 views

ParallelSubmit and WaitNext: check if a job is done before calling WaitNext

Say I process something with ParallelSubmit: LaunchKernels[] job = ParallelSubmit[1 + 2] Parallel`Developer`QueueRun[] Then, I will use ...
2
votes
0answers
67 views

What was the thought behind the parallel versions of Table, Map, etc..?

There separately named versions of the parallel analogues of Map[] and friends: ...
2
votes
0answers
103 views

Bug in parallel computing

The following is a simplified example. We construct a library function f and another library function g, in which the library function f is called. ...
2
votes
0answers
89 views

Parallel Kernel Status

For parallel computing, the window Parallel Kernel Status shows some information on the subkernels. I can use two kernels, and which each of them I can do parallel computations. However, it seems that ...
2
votes
0answers
117 views

ParallelTable kernels die - maybe Mathlink timeout?

I am new to parallel computing in Mathematica. I am using LaunchKernels[16] ParallelTable[] to parallelize calculations. The ParallelTable consists of approx ...
2
votes
0answers
36 views

do successive parallel computations require distribute definitions at each intermediate step?

Suppose I am doing a parallel computation, say, using parallel table. I distribute necessary definitions to all kernels. After computation is done, I would like to use the first result in another ...
2
votes
0answers
269 views

ParallelSum in Mathematica 9 vs 10

Here's part of a package I'm working on: ...
2
votes
0answers
210 views

Death of parallel sub-kernels

EDIT : Finally, the new Mathematica 10.0 seems to fix it. I have a little parallelization problem. I wrote a code to generate a quasicrystal by dynamical generation. Here the code: ...
2
votes
0answers
104 views

ParallelTable and expensive functions

I would like to build up a 2D-Table using ParallelTable as follows ...
2
votes
0answers
383 views

Remote kernel shuts down after several seconds

I have Mathematica 9 installed on two Windows Server 2008 R2 machines. Now i want to try using one of them as a frontend with local kernels and the other one for remote kernels. OpenSSH 6.5 for ...
2
votes
0answers
119 views

Whether NIntegrate evaluation is multithreaded or not?

About MultiThread evaluation I am talking about here, see this post I am NIntegrate the function grarm[e, 3, dimension, 1, 1, 0.005]. There is matrix operation in ...
2
votes
0answers
82 views

Setting Up Lightweight Grid On Multiple Clients Programmatically

I have a number of computers which all might need to use the same (client) configuration for the Wolfram Lightweight Grid at some point. I know that there are ways to programmatically set certain ...
2
votes
0answers
133 views

correct usage of ParallelTable

I am trying to parallelize my script. Here is a MWE of how I have done it: ...
1
vote
0answers
47 views
+50

The module load only works for the first node, not all. Is there's a way to specify the full path to the math command when launching remote kernels

Running a Mathematica .m file in cluster for parallel computing. But get an error "bash: math: command not found". Is there's a way to specify the full path to the ...
1
vote
0answers
32 views

Inconsistent font rendering when using a remote kernel on a different OS

I have a Mathematica 10.0 on Linux and one on OSX. I'm using the OSX as a remote kernel. During parallel computation, if I render some graphics, the fonts and axes are not rendered the same way ...
1
vote
0answers
42 views

How many kernels to set based on CPU?

Looks like 4 kernels are enough? My CPU is i7 4790K as given here. Does it also depend on the tasks and the codes? My understanding is that regardless what the benchmark says, I could always use ...
1
vote
0answers
57 views

ParallelMap not using all of the available kernels

Please note: I cannot provide the code here, since the project is far too big to post at any Internet forum. I have a computer with eight logical cores. I have Mathematica Version 10.0.2, and I am ...
1
vote
0answers
84 views

Avoid copying memory for each subkernel

Here is a simplified example illustrating my issue; ...
1
vote
0answers
151 views

Can GPU (CUDA) speed up LinearSolve?

Is there a way to use GPU (CUDA) to speed up LinearSolve?
1
vote
0answers
117 views

What does this Java error mean? Why is it raised?

I want to launch some kernels on a cluster that uses SGE. The server's operating system is Linux Cent-OS. According to Mathematica documentation, I've added these commands inside my code: ...
1
vote
0answers
76 views

Symbolic Nullspace computation in parallel

I am trying to determine the nullspace of a large symbolic matrix. The single core evolution seems to take too long. I tried Parallelize[] on the commands ...
1
vote
0answers
55 views

Troubles with ParallelEvaluate

I need to evaluate the same expression on all subkernels just like ParallelEvaluate does. This code distributes perfectly (2 kernels): ...
1
vote
0answers
81 views

NMaximize stops evaluating my function when run in parallel

I am trying to maximize the results of a Monte Carlo simulation with respect to some input variables. ...
1
vote
0answers
47 views

Placing package in existing Utilities` context has odd paraellelization effect

I noticed this in V.10 but possibly it was in effect in V.9. I have a package which in simplified form looks like:- ...
1
vote
0answers
155 views

Diagonalization in parallel

I would like to diagonalize one unitary non-sparse matrix of size 12870 with complex number entries (not symbols, this is really a numerical problem). Is is possible to make eigensystem run in ...
1
vote
0answers
56 views

Parallelizing an associative binary operation

I have an associative binary operation '.', which I often use in this sort of way: result = expr1 . expr2 . expr3 . expr4 The ...
1
vote
0answers
150 views

Why doesn't ParallelMap improve the performance of FoldList?

This question arose during my analysis of the following problem: filling gaps. (The code used here is Simon Woods's work.) FoldList seems to not cooperate with ...
1
vote
0answers
197 views

Parallel linear algebra with arbitrary precision

Is it possible to do parallel linear algebra with arbitrary precision within Mathematica (in a simple manner, as is done for the machine precision)?