Edric Ellis

MathWorks

Last seen: Today | Actif depuis 2011

Developer at MathWorks, working on Parallel Computing Toolbox. Professional Interests: parallel computing, distributed arrays, gpu

Statistiques

All

Afficher les badges

Content Feed

A répondu
How do you organize data from a parfor loop?
You can do this more directly by flattening the parfor loop, and using a combination of ind2sub and reshape. The idea is as foll...

8 jours il y a | 0

A répondu
How to use parallel.pool.Constant to save on overheard for large array constant across all calls to parfeval?
You've got a subtle flaw in your first piece of code that allows the program to run, but it does not actually benefit from the u...

8 jours il y a | 0

| A accepté

A répondu
Read/Write Limitations with Parallel Computing Toolbox (parfor)
Generally speaking, parfor is conservative in what it allows to ensure that by default, all variable accesses are safe. It is s...

22 jours il y a | 0

| A accepté

A répondu
Parallel workers automatically shutting down in the middle of RL parallel training.
If workers are leaving the pool one at a time while the pool is busy, this almost certainly means that they are crashing. I reco...

27 jours il y a | 0

| A accepté

A répondu
Clear persistent variables of function on worker
Whenever you issue a clear command at the client, the same clear command is issued on the workers. So you can simply do clear p...

29 jours il y a | 1

A répondu
Error using parpool. Parallel pool failed to start with the following error. For more detailed information, validate the profile 'Processes' in the Cluster Profile Manager.
This is probably the following known problem in R2023a: https://www.mathworks.com/support/bugreports/2968710 . There are workaro...

environ 2 mois il y a | 0

A répondu
Matlab spmd Collecting Composite Arrays
You missed a crucial clue from your question - where are you assigning hx. I'm going to guess you have something like this: spm...

3 mois il y a | 0

| A accepté

A répondu
Matlab spmd and CoolProp (CPython library)
You need to avoid transferring the objects to the workers, and instead simply build the objects directly on the workers. The sim...

3 mois il y a | 0

| A accepté

A répondu
How to shut down all running workers of paarpools?
You can shut down all remaining workers of the currently running pool by executing: delete(gcp('nocreate')) There should be no...

3 mois il y a | 0

A répondu
Assigning values to a vector out of order in parallel (parfor)
One of the constraints of parfor is that indexed outputs like myVec must be "sliced" - that means that you must assign only elem...

3 mois il y a | 0

| A accepté

A répondu
Is it possible to create multiple parallel pools simultaneously
There's no direct way to do what you're after here. (As @Walter Roberson points out, you can have a backgroundPool and a parpool...

4 mois il y a | 1

| A accepté

A répondu
Cluster multi-gpu training Error: Current pool is not local.
I think you need to specify ExecutionEnvironment="parallel" for this situation. According to the trainingOptions reference page,...

5 mois il y a | 0

| A accepté

A répondu
Is is possible to run a batch job on an MJS Cloud Center cluster with SpmdEnabled set to false?
Unfortunately, this option is not supported at the moment for batch jobs. I realise it's probably rather a big change to your co...

5 mois il y a | 0

| A accepté

A répondu
I need help regarding parfeval implementation with object handle. I have created an object handle, the member functions of which i want to run as parallel processes.
The documentation here applies to parfeval as well as parfor in the sense that the object instances on the workers are not conne...

6 mois il y a | 0

| A accepté

A répondu
How to avoid broadcast variable while optimizing a cost function in parallel computing?
Investigating performance of parfor loops can be a bit tricky. Here are a few pointers: Do you happen to know if your function ...

6 mois il y a | 0

A répondu
parfor with temporary variable worning
I think there are a couple of problems here. Firstly, I think you need to change the way you're assigning into dcm_stack because...

6 mois il y a | 0

| A accepté

A répondu
Can parfor be used if no Parallel Computing Toolbox licenses are available?
The implementation of parfor is intended to gracefully fall back to running in serial if a Parallel Computing Toolbox licence ca...

7 mois il y a | 0

A répondu
Slicing variable in parfor loop
In this case, you don't actually have a data dependency between the loop iterations. You can use a parfor reduction to do this, ...

7 mois il y a | 0

| A accepté

A répondu
GPU0 and GPU1 in matlab?
GPU Computing in MATLAB supports only NVIDIA devices, so your Intel graphics card cannot be used by gpuArray.

7 mois il y a | 1

| A accepté

A répondu
Sliced Variables in N Choose K Search
This is a limitation of the parfor analysis when it comes to indexing sliced variables in a nested for loop. There is another wo...

7 mois il y a | 0

| A accepté

A répondu
Use Parfoor Loop for Parameter Sweep Optimization
You can do this using a custom reduction function in parfor. Like this: p = 1:10; someFcn = @sin; % Use a 2-element vector to...

7 mois il y a | 2

| A accepté

A répondu
SMPD block getting stuck and not executing
I suspect the problem here is that the way that you're using spmdSend and spmdReceive doesn't quite match how they work. When yo...

7 mois il y a | 0

| A accepté

A répondu
How can I Ensure Transparency in parfor-Loops when using cell2table ?
Most of the table-building functions can have this problem if you do not specify 'VariableNames', like this (tested in R2019a): ...

7 mois il y a | 2

A répondu
Parrallel run using parfor and institute server?
The problem with converting the outer loop to parfor is described in the error message that you get when you try: Error: When i...

8 mois il y a | 0

A répondu
Incompatible indexing of variable for parfor
The total size in bytes of Network is 5*12*76*8 = 36480. That is tiny. Ignore the parfor "broadcast" warning - that is intended ...

8 mois il y a | 0

A répondu
Minimum/maximum CUDA Toolkit requirement for R2022a
To run code using gpuArray, you do not need to install a CUDA Toolkit. You only need a CUDA driver. You should always* install t...

9 mois il y a | 0

A répondu
Is there an easy way to find out which workers are running on the same host in a Generic Cluster job so I can efficiently allgather?
You can use gop to perform general MPI-style all-reduce operations, and a special case of that is gcat which can operate as an M...

9 mois il y a | 0

| A accepté

A répondu
Unsure why parfor loop does not work in this case
The problematic variables appear to be both M and P. For M, you have multiple different "reads" from the variable after the init...

10 mois il y a | 0

A répondu
Writing to .mat simultaneously with parfor loop
Don't attempt to save from multiple workers to the same file, this will sometimes fail (exactly when/how it fails depends on you...

10 mois il y a | 2

| A accepté

A répondu
SPMD to process different iteration on different worker
It's not entirely clear to me quite what you're after. You might want for (drange) which spreads the iterations of a for loop ov...

10 mois il y a | 0

Charger plus

Edric Ellis

MathWorks

Statistiques

File Exchange

File Exchange

File Exchange

File Exchange

MATLAB Answers

MATLAB Answers

MATLAB Answers

MATLAB Answers

MATLAB Answers

MATLAB Answers

Cody

Content Feed