Subscribe

UiPath AI Center

UiPath AI Center

AI Center Single Node

This page details the hardware, software requirements as well as prerequisites for installing AI Center Single node.

Hardware Requirements

The table below makes some recommendations averaged across generic models (small and large).

CPU

RAM (GB)

OS/Boot Disk (GB)

External Data Disk (GB)

Models Served

Concurrent Models Trained

8

52

200

500

3

1-2

12

64

200

1000

3-4

2

For improved performance, we recommend using a SSD drive for both primary and secondary disks.

❗️

Airgapped installation

An airgapped installation requires a 500 GB for OS/Boot Disk.

Sizing the machine

There is no universal value for how much resources one ML Skill/Pipeline job will consume as it depends of the model. However, here are minimum resources used by an ML Skill/Pipeline job along with resources used by a UiPath Document Understanding Model as a baseline.

Use

CPU

RAM (GB)

GPU

Core Services

4.7

16

Minimum for Serving (ML SKill)

0.6

2

Minimum for Training (Pipeline)

1

4

DU model Serving

1

4

DU model Training

2

24

1

🚧

Before provisioning a machine, be sure to read the installation instructions. The external data disk attached to the machine must be un-formatted and must be of type disk not partition. See step 1. Provision a Machine.

GPU Requirements

Only NVIDIA GPUs are currently supported. If you have constrains on model training time, it is recommended you add a GPU with at least 11 GB of Video RAM. GPU is always required for DU model training. If you run UiPath OCR (the non-Edge version), on AI Center, to process more than 2 million pages per year, GPU is highly recommended for better performance.
You are responsible for installing GPU drivers before you can use GPU in AI Center. For more information, see Prerequisites for Installation below.

Software Requirements

Operating System

The following table lists the operating system(s) officially supported for the AI Center on-premises installation.

OS

Version

Ubuntu

18.04 LTS

RHEL

7.4, 7.5, 7.6, 7.7, 7.8, 7.9

CentOS

7.4, 7.5, 7.6, 7.7, 7.8, 7.9

  • Machine should have lvm2 installed.
  • OverlayFS storage drivers for docker should be usable, you can check prerequisites on docker documentation. No need to install anything just fulfill the prerequistes.

Browsers

The following table lists the browser(s) officially supported for the AI Center on-premises installation.

Browser

Version

Google Chrome

64 or above

Microsoft Edge

80 or above

Mozilla Firefox

66 or above

Prerequisites for Installation

Before starting the UiPath installation the following prerequisites are needed:

  • Orchestrator 20.4.3 (or higher)
    See the guide here for various ways to install Orchestrator.
  • SQL Server 2014 (or higher)
    It is highly recommended that you use the same SQL Server as was used when installing Orchestrator as detailed here. For the installation, you will require the hostname, admin username, and password of this SQL Server.

🚧

Make sure that SQL Server Authentication mode is enabled.

📘

Note

AI Center uses SQL solely for metadata storage. This means that the amount of data store is very small. There is no need to provision a lot of storage capacity for these tables.

  • GPU Prerequisites
  1. Run nvidia-smi to check if the driver is installed.
  2. Run /usr/bin/nvidia-container-runtime to check if the container runtime is installed.

AI Center Architecture

AI Center runs on a kubernetes cluster. All communication into and out of the cluster is secured with HTTPS (TLS). Tenant and user-specific traffic uses modern protocols (OAuth2.0 and OpenID) supported by UiPath's Identity Server.

The diagram below shows a detailed architecture diagram of the various components in AI Center.

At a high-level, AI Center core services manage the deployment and training of machine learning models.

A deployment of a machine learning model (called an ML Skill) is a container with the code and model artifacts. AI Center creates an end-point from that container that is permissioned and replicated.

A training or evaluation of a machine learning model will create a container image on-the-fly and execute code predefined in by the AI Center user or by an out-of-the-box retrainable model.

Updated 4 days ago


AI Center Single Node


Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.