There are multiple options for scaling your MATLAB® programs and Simulink® models to the cloud. Take advantage of your desktop resources using Parallel Computing Toolbox™, and then easily scale your application beyond the desktop to use additional hardware resources without changing your algorithmic code.
Try your deep learning applications with multiple high-end GPUs on Amazon EC2. To help you get started, download a white paper that outlines a complete workflow.
|Parallel Computing Toolbox||MATLAB Parallel Cloud||MATLAB Distributed Computing Server for Amazon EC2||MATLAB Distributed Computing Server - Private Cloud|
|Maximum Workers*||No limitation**||16||256||No limitation**|
|Hardware Resources||Desktop computer||MathWorks Cloud||Amazon EC2 instances (billed by Amazon Web Services)||Private cloud, other cloud services, on premise and ad-hoc clusters, and grids.|
|First-Time Configuration Effort||None||A few clicks in MATLAB||Amazon EC2 sign up and set up followed by a few clicks in Cloud Center and MATLAB.||Software installation followed by scheduler configuration.|
|Time to Access Configured Solution||Instant||<90 seconds||<15 minutes||Solution dependent|
|Customization Options||None||None||Available through Cloud Center. Options include cluster size, machine type, storage options.||Options include multiple cluster configurations, storage types, and schedulers.|
|Licensing Model||Toolbox license||Self-serve On-demand license||On-demand, perpetual or term license.||On-demand, perpetual or term license.|
|Geographic Availability||Worldwide||United States and Canada||United States, Canada, and other select countries.||Worldwide|
* MATLAB computational engines.
** Recommend one worker per physical core.