Enterprise Deployment with MATLAB Production Server

Run MATLAB^® code on a server and access it using various clients

MATLAB Production Server™ lets you run MATLAB functions on a server and access them via client applications. You create applications in MATLAB, package them using the Production Server Compiler app or the compiler.build.productionServerArchive function in MATLAB Compiler SDK™, and deploy them to the MATLAB Production Server. Access deployed packages through client applications written in Java^®, .NET, Python^®, C, or C++ using client-specific libraries, or through HTTP/HTTPS requests using the RESTful API.

For information on how to write client applications using the RESTful API, Java, .NET, Python, C, and C++, see Client Programming (MATLAB Production Server).

Apps

Production Server Compiler

Package MATLAB programs for deployment to MATLAB Production Server

Functions

expand all

App Launcher

deploytool Open a list of application deployment apps

Deployable Archive

`compiler.build.productionServerArchive`	Create an archive for deployment to MATLAB Production Server or Docker (Since R2020b)
`compiler.build.ProductionServerArchiveOptions`	Options for building deployable archives (Since R2020b)
`compiler.build.excelClientForProductionServer`	Create Microsoft Excel add-in for MATLAB Production Server (Since R2021b)
`compiler.build.ExcelClientForProductionServerOptions`	Options for building Excel add-ins (Since R2021b)
`compiler.package.excelClientForProductionServer`	Create an installer for an Excel client for MATLAB Production Server (Since R2021b)
`compiler.package.ExcelClientForProductionServerOptions`	Options for creating an installer for an Excel client for MATLAB Production Server (Since R2021b)
`compiler.build.Results`	Compiler build results object (Since R2020b)
`mcc`	Compile MATLAB functions for deployment

RESTful API JSON Encode and Decode

`mps.json.encode`	Convert MATLAB data to JSON text using MATLAB Production Server JSON schema
`mps.json.decode`	Convert a character vector or string in MATLAB Production Server JSON schema to MATLAB data
`mps.json.encoderequest`	Convert MATLAB data in a server request to JSON text using MATLAB Production Server JSON schema
`mps.json.decoderesponse`	Convert JSON text from a server response to MATLAB data

RESTful API Metrics

`prodserver.metrics.incrementCounter`	Create Prometheus `counter` metric (Since R2022a)
`prodserver.metrics.setGauge`	Create Prometheus `gauge` metric (Since R2022a)

Persistence

Controller

`mps.cache.control`	Create a persistence service controller object
`start`	Start a persistence service and attach it to a MATLAB session
`stop`	Stop a persistence service and detach it from a MATLAB session
`restart`	Restart a persistence service and attach it to a MATLAB session
`attach`	Connect MATLAB session to persistence service that is already running
`detach`	Disconnect MATLAB session from persistence service that is already running
`ping`	Test whether the persistence service is reachable
`version`	Version number for persistence provider

Cache

`mps.cache.connect`	Connect to cache, or create a cache if it doesn't exist
`bytes`	Return the number of bytes of storage used by value stored at each key
`clear`	Remove all keys and values from cache
`flush`	Write all locally modified keys to the persistence service
`get`	Fetch values of keys from cache
`getp`	Get the value of a public cache property
`isKey`	Determine if the cache contains specified keys
`keys`	Get all keys from cache
`length`	Number of key-value pairs in the data cache
`purge`	Flush all local data to the persistence service
`put`	Write key-value pairs to cache
`remove`	Remove keys from cache
`retain`	Store remote keys from cache locally or return locally stored keys

Mutex

`mps.sync.mutex`	Create a persistence service mutex
`acquire`	Acquire advisory lock on persistence service mutex
`own`	Check ownership of advisory lock on a persistence service mutex object
`release`	Release advisory lock on persistence service mutex

Objects

expand all

Persistence

`mps.cache.Controller`	Manage the lifecycle of a persistence service in a MATLAB testing environment
`mps.cache.DataCache`	Represent cache concept in MATLAB code
`mps.sync.TimedMATFileMutex`	Represent a MAT-file persistence service mutex
`mps.sync.TimedRedisMutex`	Represent a Redis persistence service mutex

Topics

Deployable Archive Creation

Create Deployable Archive for MATLAB Production Server
Learn how to package a MATLAB function into a deployable archive for MATLAB Production Server.
Package Deployable Archives from Command Line
Package MATLAB applications into deployable archives from the command line using the mcc command.

MATLAB Programming to Support In-Memory Caching

Data Caching Basics
Set up the server to use Redis™, start a persistence service, and create a data cache.
Manage Application State in Deployed Archives
Test application before deployment and cache data in application archives deployed to MATLAB Production Server.

MATLAB Programming to Support Custom Metrics

Create Custom Prometheus Metrics
This example shows how to create custom Prometheus^® metrics on the server and retrieve them using the metrics service.

MATLAB Programming Guidelines

MATLAB Coding Guidelines
Best practices for writing MATLAB code when deploying to MATLAB Production Server.
State-Dependent Functions
Dealing with MATLAB functions that carry state.
Deploying MATLAB Functions Containing MEX Files
How to include MEX files when deploying MATLAB functions.
Supported MATLAB Data Types for Client and Server Marshaling
List of supported and unsupported MATLAB data types.
Modifying Deployed Functions
How to modify and redeploy MATLAB functions.

Client Applications

Create MATLAB Production Server Java Client Using MWHttpClient Class (MATLAB Production Server)
Define an interface, instantiate a static proxy object, and then call the deployed function.
Create a C# Client (MATLAB Production Server)
Learn how to call a MATLAB function deployed to MATLAB Production Server from a C# application.
Create a Python Client (MATLAB Production Server)
Learn how to call a MATLAB function deployed to MATLAB Production Server from a Python application.
Create a C++ Client (MATLAB Production Server)
Learn how to call a MATLAB function deployed to MATLAB Production Server from a C++ application.

Testing

Test Client Data Integration Against MATLAB
Test the integration code used by your client against MATLAB.
Test Web Request Handlers
Configure access to the routes file in the Production Server Compiler app, then test MATLAB functions for deployment to MATLAB Production Server.

Excel Add-Ins

Create and Install Deployable Archive with Excel Integration for MATLAB Production Server
Learn how to create a deployable archive with Excel^® integration for MATLAB Production Server using a MATLAB function.

Parallel Computing

Use Parallel Computing Resources in Deployable Archives
Pass a cluster profile to a deployable archive that uses Parallel Computing Toolbox™ or MATLAB Parallel Server™.

Troubleshooting

Testing

MATLAB Not Responding to Web Requests Made to Test Server
Troubleshoot why MATLAB does not respond to web requests made from MATLAB to a local test server.

Excel Add-In

XLA File Not Generated
Solution to .xla file not being generated.
Server Configuration Add-in Not Enabled
Solution to enable Server Configuration add-in.
Error Using a Variable Number of Outputs
Solution to support variable number of outputs.