Gurobi Instant Cloud

Fully managed Gurobi Solver on Microsoft Azure® and Amazon Web Services®

Great for…

  • Handling spikes in demand
  • Solving challenging models
  • Meeting periodic optimization needs
  • Delivering cloud-based solutions
  • Providing cloud-based failover

Easy and Robust

  • Automatically start, manage and stop multiple machines
  • Access from your existing applications
  • Select dedicated machines from a data center near you on AWS or Azure
  • Stay secure with data encryption

Cost Effective

  • Use and pay for only what you need
  • Reduce or eliminate local data center costs
  • Support Windows, Linux and Mac clients
  • Access includes Gurobi Support

Cloud Manager

The Instant Cloud Manager lets you manage your licenses, machines, pools and settings. You can also access the active machines and the history of recently used machines. The licenses can be shared among users of the same organization so that multiple users can manage a set of machines and pools.


To learn more, you can follow the steps from the getting started tutorial. To open the Instant Cloud Manager, you must be logged in and have an Instant Cloud license.


A pool defines a configuration (machine type, region and other parameters) as well as a number of compute servers and distributed workers. Instant Cloud provides you with predefined pools for your licenses, and you can create new ones depending on your needs, for example:


If you see that your application needs a larger machine or more machines, just update the pool definition. The next time the application will access the pool, the new machines will be launched according to the updated specification.

With the v8 engine, pools can be scaled up or down by starting or stopping machines of a pool in matter of minutes.

Regions and Machine types

Gurobi Instant Cloud is available in 5 AWS regions (California, Virginia, Sydney, Tokyo and Frankfurt) and 3 Azure regions (US East, US West 2, EU West). Depending on the Cloud provider and region, you are able to select the right configuration for your application. Note that some configurations may not be available in all regions.

Provider Type RAM (GB) #Cores Category
AWS c5.4xlarge 32 8 Compute Optimized
c5.9xlarge 72 18 Compute Optimized
c5.18xlarge 144 36 Compute Optimized
r4.8xlarge 244 16 Memory Optimized
r4.16xlarge 488 32 Memory Optimized
r5.24xlarge 768 48 Memory Optimized
z1d.12xlarge 384 24 High Performance
Azure F16v2 32 8 Compute Optimized
F32v2 64 16 Compute Optimized
F72v2 144 36 Compute Optimized
E64v3 432 32 Memory Optimized
H16m 224 16 High Performance

Automated Provisioning

Your pools are integrated with the Gurobi client libraries so that machines will be launched automatically when you start an optimization session in your application. If the machines are already launched, they will be immediately used for optimization. You can select the pool you want to use by downloading the license file associated with each pool. No need to know more about cloud computing, Instant Cloud will do it for you.

$ gurobi_cl afiro.mps
Waiting for cloud server to start..........
Capacity available on '999999-default' cloud pool - connecting...
Established HTTPS encrypted connection

Gurobi Optimizer version 8.1.0
Copyright (c) 2018, Gurobi Optimization, LLC


Solved in 3 iterations and 0.74 seconds
Optimal objective -4.647531429e+02

The cloud license file will work seamlessly with all languages or environments supported by Gurobi: C++, Python, MATLAB, Java, .Net, C or R. You can also specify directly in your program the access id and the secret key found in the license file using the Gurobi environment constructor (see the Gurobi reference manual).

Job Management and History

With the v8 engine, you can now access the list of queued, running or recently ended jobs. The optimization expert can then open the job dashboard to display various views about the selected job:


In additon, jobs can be automatically archived for later access to the metadata and the engine log.

Active Machines and History

As we have shown, the machines will be started when needed by Instant Cloud using the pools. In addition, you can list the active machines or access the history of recent machines to monitor your environment.


Advanced Integration

Using the Instant Cloud REST API, you can build custom frameworks to manage machines and pools. You can create pools, start and stop machines or pools without dealing with the details of the cloud infrastructure. The API is secured by encrypted communications (HTTPS) and API keys that you can retrieve from the Instant Cloud Manager.