System Requirements

• Supported Operating Systems

• Hardware Requirements

• Supported Queueing Systems

• Extra considerations for visualization and interacting with Maestro

• GPGPU Requirements

• System requirements for individual products

Supported Operating Systems

• Windows
• Mac
• Linux

sudo yum/dnf install <lib>

sudo yum/dnf install <lib>

sudo apt-get install <lib>

Hardware Requirements

Supported Queueing Systems

To run jobs on a remote host, a queueing system is required. The following queueing systems are supported:

• PBS Pro
• Grid Engine, including SGE, Open Grid Scheduler, and Univa GE, minimum version 6.2
• LSF, minimum version 7.0.2
• Slurm, minimum version 2.1 (minimum version 18.08.2 for remote license checking, see Configuring Remote License Checking)
• Torque

Supported Features

See Setting Up License Checking for Queueing Systems for more information on setting up license checking.

Extra considerations for visualization and interacting with Maestro

GPU

We recommend using a graphics card that supports hardware-accelerated OpenGL with at least 1GB onboard memory and an up-to-date vendor-supplied graphics driver.

Network file share

Using a networked file share mounted via CIFS (Samba) is not recommended, as Maestro projects use SQLite databases that have locking dependencies not typically available on them.

Mouse

We recommend using a 3-button mouse with a scroll wheel.

3D stereo display

For a 3D stereo display, we recommend Looking Glass Monitors.

Remote display

A local installation of Maestro on your laptop or workstation will always run better than running it from a remote compute resource. If you need to run Maestro remotely, various protocols exist for virtual desktops with varying degrees of compatibility with OpenGL, Qt, and other graphics dependencies that Maestro has. Support for such protocols is outside of our control.

We strongly recommend against running Maestro via basic X11 forwarding, e.g., via “ssh -X workstation”, as the performance will be poor.

Coupling an up-to-date version VirtualGL (www.virtualgl.org) with a remote desktop protocol can make a remote session almost as responsive as a local environment. Once installed, you will launch Maestro as:

vglrun $SCHRODINGER/maestro

Typical “compute nodes” on HPC environments often do not have high quality graphics cards and are not optimal for running Maestro; their graphics hardware is often dedicated to GPU computing.

GPGPU Requirements

(General-purpose computing on graphics processing units)

Listed here are the GPU computing requirements for Desmond, Deep AutoQSAR, FEP+, GPU Shape, and WaterMap.

Requirements for Jaguar, Active Learning and other products can be found on their individual pages below.

We support the following NVIDIA solutions:

Deprecated

• Support for the Tesla M40 and M60 cards is deprecated.

Notes

• We support only the NVIDIA 'recommended / certified / production branch' Linux drivers for these cards with minimum CUDA version 12.0.

• For information on pre-configured Schrödinger compatible GPU boxes see MD Compatible Systems and FEP+ Compatible Systems.

• Standard support does not cover consumer-level GPU cards such as GeForce GTX cards.

• In the 2024-2 release, the "Discounted Cores" of an L4 was decreased from 7424 to 5120. This adjustment was made to align with the L4's status as a preferred card due to its widespread availability on-premises and through cloud providers. Additionally, the L4's low power consumption, speed, and sufficient memory make it suitable for running the majority of workflows in a configuration with one GPU per node.

• If you already have another NVIDIA GPGPU and would like to know if we have experience with it, please contact our support at help@schrodinger.com.

System requirements for individual products