Skip to main content

Reveal Review Publication

System Requirements
Table 1. Reveal AI Version 3.10+ System Recommendations[1]

Component Name

Description

Requirements

SQL Server

  • Host NexLP Admin and StoryBook databases

  • Hardware

    • Processors: 4 core or more

    • Memory: 64 GB or higher

    • 1 TB free space or more

    • SSD highly recommended

  • Software

    • Windows Server 2012 and up

    • SQL Server SQL 2016 SP2 or above, Standard/Enterprise SP1/SP2 highly recommended

    • SQL Full Text enabled

Analytics Processing Server[2]

  • Host Reveal AI processing service

  • Build Reveal AI analytics index

  • Hardware

    • 8 core or more

    • 64 GB min

    • 500 GB free disk space

  • Software

    • Windows Server 2012 and up

    • SQL Management Studio (2016 SP2 and up)

    • .NET Version 4.8 or higher

    • JRE 1.8, 64bit

AI Server (optional)

  • Host AI services, including COSMIC, Alerts, Language Modelling, Vectors

  • Hardware

    • Processors: 4 core

    • Memory: 64 GB  min, 128 GB highly recommended3

    • 500 GB free disk space

  • Software

    • Windows Server 2012 and up

    • JRE 1.8, 64bit

    • .NET Version 4.8

Web Server

  • Hosts the Reveal AI web pages

  • Hardware

    • Processors: 4 core

    • Memory: 8 GB  min

    • 20 GB free disk space

  • Software

    • Windows Server 2012 and up

    • IIS 8 or higher, Asp.net enabled

    • .NET Version 4.8, ASP.net enabled

    • Visual C++ Redistributable for Visual Studio 2015 (x64) + (x86)

    • SSL Certificate (optional)

    • URL Rewrite is enabled

Caching Server[4](optional)

  • Hosts caching service

  • Hardware

    • Processors: 2 core

    • Memory: 8 GB min

    • 20 GB free disk space

  • Software

    • Windows Server 2008 R2 and up

File Server (optional)

  • Host processing output

  • Hardware

    • Processors: 2 core

    • Memory: 8 GB min

    • 1TB free disk space

  • Software

    • Windows Server 2008 R2 and up

3rd party integration

  • Platform Integration

  • Relativity®  version 10.3 required if running integrated mode

  • RevealTM  version 9+

Client side

  • Interact with Reveal AI

  • Google Chrome



Note

[1] Depending on implementation and project load, these servers may be stacked into one box configurations.

[2] For one language, estimate 500k to 1M documents analyzed and written to database per day, per analysis processing server box.

[3] AI Server may need 4 GB memory per 1 million documents during active COSMIC analysis. If using High Precision Classification (HPC), recommend increasing Memory to 64 or 128 GB.

[4] By default caching server uses port 6379, make sure other servers can access this port on caching server.

User Accounts Requirements

Account Name

Description

Permissions

NexLP Service Account

  • Creates new SQL databases for each case

  • Used by front-end to connect to SQL Server

  • Runs Processing/AI Services

  • Set up Web Application Pool on Web Server

  • Is a domain user

  • Has Read/Write permission to NEXLPSHARE folder (located in network), and service logs folder

  • Has Read permission to source text files located in network

  • Is an Admin on NexLP SQL server

  • Has read/write permission to source SQL database, or read permission (if “update source database on processing import” is disabled)

  1. The NexLP Service Account is used to set up IIS application pool on the Web Server. If the attribute “manualGroupMembership” for processModel is set to “true”, this user will need to be added to the IIS_USERS group, see the link below for more details:

  2. The common language runtime (CLR) integration feature is off by default. Check MSDN to enable it for the output SQL server, refer to the MSDN article here:

In this section: