Clusters and Clouds

Discover cluster resources, and work with cluster profiles

If your computing task is too big or too slow for your local computer, you can offload your calculation to a cluster onsite or in the cloud to run your MATLAB^®code with minimal changes. TryParallel>Discover Clustersin the MATLAB toolstrip to find out if you already have a cluster available.

If you already have a cluster with a scheduler, you can integrate MATLAB with it usingMATLAB Parallel Server™. Alternatively, if you do not have an existing scheduler, thenMATLAB Parallel Serverprovides MATLAB Job Scheduler.

Functions

expand all

Key Functions

`parcluster`	Create cluster object
`parpool`	Create parallel pool on cluster
`gcp`	Get current parallel pool
`shutdown`	Shut down cloud cluster
`start`	Start cloud cluster
`wait (cluster)`	Wait for cloud cluster to change state
`parallel.defaultClusterProfile`	Examine or set default cluster profile
`parallel.exportProfile`	Export one or more profiles to file
`parallel.importProfile`	Import cluster profiles from file
`saveProfile`	Save modified cluster properties to its current profile
`saveAsProfile`	Save cluster properties to specified profile
`pctconfig`	Configure settings forParallel Computing Toolboxclient session

Classes

expand all

Key Classes

`parallel.Pool`	Parallel pool of workers
`parallel.Cluster`	Access cluster properties and behaviors
`pctRunOnAll`	Run command on client and all workers in parallel pool

Examples and How To

Cluster Setup

Discover Clusters and Use Cluster Profiles
Find out how to work with cluster profiles and discover cloud clusters.
Scale Up from Desktop to Cluster
This example shows how to develop your parallel MATLAB® code on your local machine and scale up to a cluster.
Process Big Data in the Cloud
This example shows how to access a large data set in the cloud and process it in a cloud cluster using MATLAB capabilities for big data.
Benchmark Your Cluster with the HPC Challenge
This example shows how to evaluate the performance of a compute cluster with the HPC Challenge Benchmark.

Deep Learning

Scale Up Deep Learning in Parallel, on GPUs, and in the Cloud(Deep Learning Toolbox)
Options for deep learning with MATLAB in parallel and using multiple GPUs, locally or in the cloud.
Deep Learning with MATLAB on Multiple GPUs(Deep Learning Toolbox)
Speed up deep neural network training using multiple GPUs locally or in the cloud.
Train Network Using Automatic Multi-GPU Support(Deep Learning Toolbox)
This example shows how to use multiple GPUs on your local machine for deep learning training using automatic parallel support.
Use parfor to Train Multiple Deep Learning Networks(Deep Learning Toolbox)
This example shows how to use aparforloop to perform a parameter sweep on a training option.
Use parfeval to Train Multiple Deep Learning Networks(Deep Learning Toolbox)
This example shows how to useparfevalto perform a parameter sweep on the depth of the network architecture for a deep learning network and retrieve data during training.
Train Deep Learning Networks in Parallel(Deep Learning Toolbox)
This example shows how to run multiple deep learning experiments on your local machine.
Train Network in Parallel with Custom Training Loop(Deep Learning Toolbox)
This example shows how to set up a custom training loop to train a network in parallel.
Upload Deep Learning Data to the Cloud(Deep Learning Toolbox)
This example shows how to upload data to an Amazon S3 bucket.
Send Deep Learning Batch Job to Cluster(Deep Learning Toolbox)
这个例子展示了如何发送深度学习培训ing batch jobs to a cluster so that you can continue working or close MATLAB during training.

Concepts

Specify Your Parallel Preferences
Specify your preferences, and automatically create a parallel pool.
Plugin Scripts for Generic Schedulers
How to use plugin scripts to set up generic schedulers.
Set Environment Variables on Workers
Copy system environment variables from the client to workers in a cluster.

Related Information

Deep Learning in Parallel and in the Cloud(Deep Learning Toolbox)
Get Started with MATLAB Parallel Server(MATLAB Parallel Server)
Scaling to clusters and cloud

Featured Examples

Scale Up from Desktop to Cluster

Develop your parallel MATLAB® code on your local machine and scale up to a cluster. Clusters provide more computational resources to speed up and distribute your computations. You can run your code interactively in parallel on your local machine, then on a cluster, without changing your code. When you are done prototyping your code on your local machine, you can offload your computations to the cluster using batch jobs. So, you can close MATLAB and retrieve the results later.

Open Live Script

Process Big Data in the Cloud

Access a large data set in the cloud and process it in a cloud cluster using MATLAB capabilities for big data.

Open Live Script