graph-studio
Installing Graph Lakehouse
Follow the appropriate instructions below to review the prerequisites and install Graph Lakehouse on a single server or cluster.
- Review the Prerequisites
- Install Graph Lakehouse on a Single Server
- Install Graph Lakehouse on a Cluster
Review the Prerequisites
Before installing Graph Lakehouse, double-check that the following requirements are met. If any of the items are incomplete, complete those tasks before starting the installation:
- All host servers meet the requirements in Graph LakehouseGraph Lakehouse Requirements.
- The shared file system is mounted to the host servers and meets the requirements in Platform Shared File Storage Requirements.
- The prerequisite software, described in Complete the Pre-Installation Requirements, is installed.
To streamline the installation and configuration of Graph Lakehouse, Altair recommends that you run the installer with sudo privileges. In Sudo Mode, the installer can configure and start systemd services for the Graph Lakehouse processes. The services configure the appropriate resource limits (ulimits) and set $JAVA_HOME to point to the OpenJDK installation. During the installation, you are prompted to provide the platform service user name. The provided name is used as the User value in the service units, and file system permissions for the installation directory and Graph Lakehouse-generated files are also based on the specified user.
Install Graph Lakehouse on a Single Server
Follow the steps below to install Graph Lakehouse on a single server.
Download the installer to the Graph Lakehouse host server. The installer is an interactive shell script that prompts you to choose configuration options for the deployment.
Change directories to the location where you copied the script and run the following command to make the script executable:
chmod +x <script_name>
For example:
chmod +x anzograph_linux_3_1_0_r202311010839.sh
Run the following command to start the installation wizard:
sudo ./<script_name>.sh
The installer first verifies that the prerequisites listed in Prerequisite Software are installed and presents a message if any are missing. For example, on a server where OpenJDK and binutils are installed but the glibc-devel package is missing, the installer displays the following message and the installation is canceled. In this case, the user would follow the instructions in GNU C Devel Library and then restart the installation.
Starting Installer ... Prerequisite software packages
The following packages are missing on your system and are required to be installed before proceeding with the Graph Lakehouse installation:
- glibc-devel
Canceling the installation, install the missing software, and then run the installer again.
If the prerequisite software is installed, but the C++ dependencies are not, the installer presents an informational message to let you know the dependencies are required but it is recommended that you install them after the Graph Lakehouse installation. For example:
Starting Installer ... Prerequisite software packages
INFO: The following C++ dependencies are also required and missing from your system. However, Altair recommends that you install them after the Graph Lakehouse installation is complete. Follow the Post-Installation instructions in the online documentation. libarchive libarmadillo12 libboost_filesystem1_80_0 libboost_iostreams1_80_0 libboost_system1_80_0 libflatbuffers2 libhdfs3 libnfs13 libserd-0-0 libsmb2 shadow-utils
If all dependencies are installed or you proceed with the installation after getting the informational message, the license agreement is displayed. Press Enter to scroll through the text. At the end you are prompted to accept the agreement.
Starting Installer ... Please read the following License Agreement. You must accept the terms of this agreement before continuing with the installation.
ANZOGRAPH(R) DB END USER LICENSE AGREEMENT ... I accept the agreement Yes [1], No [2]
Enter 1 to accept the license agreement and continue the installation. The installer prompts you to specify the directory where Graph Lakehouse should be installed.
Where should Graph Lakehouse be installed? [/opt/cambridgesemantics]
A number of subdirectories and an uninstall script are created inside the directory that you specify. One subdirectory is named
anzographand includes the installation files. Because an anzograph directory will be created, you may not want to specify/opt/anzographas the install location because that will result in an/opt/anzograph/anzographdirectory.Press Enter to select the default installation directory or specify an alternate location and then press Enter.
Next, the wizard prompts you to specify the installation type: single standalone server, leader server, or compute server:
Type of server being installed. Server Installation Type Standalone [1, Enter], Cluster Leader [2], Cluster Compute/Worker [3]
Press Enter to accept the default option, Standalone (1).
The next prompt asks you to create the username for the administrator. This username will be specified when creating the connection between Graph Studio and Graph Lakehouse.
Set up the Graph Lakehouse Admin user. Graph Lakehouse Admin user [admin]
Press Enter to accept the default username, admin, or type another username (case-sensitive) and press Enter. The installer asks you to create a password for the Admin user.
Graph Lakehouse Admin password
Enter the case-sensitive password for the Admin user and then press Enter. It is recommended that you create a non-trivial secure password that includes some combination of upper and lower case letters and digits. For security purposes, the installer does not echo the characters that you type.
There is no limitation placed on the length of the password. However, some special characters, such as
$and*, are treated as parameters in bash. When typing the password, avoid special characters. For more information, see Quoting in the Bash Reference Manual.Next, the installer asks if the installation will be used with Graph Studio:
Is this Graph Lakehouse installation intended for use with Graph Studio? Yes [y, Enter], No [n]
Press Enter for Yes. Answering yes configures Graph Lakehouse to use the settings that are optimal for Graph Studio. Answering no would configure the settings that are optimal for Graph Lakehouse standalone use without Graph Studio.
The installer asks if you want the admin user to be able to run queries.
Do you want to use admin user to run queries? Yes [y, Enter], No [n]
Press Enter for Yes. Answering yes is required for use with Graph Studio.
Next, the installer prompts you to specify the platform service user name. The default name is
anzo.Setup Graph Lakehouse service user name Graph Lakehouse service user [anzo]
Press Enter to accept the default name,
anzo, or type an alternate name and then press Enter if a different name is used for the account.Next, the installer asks if you want to add a specific configuration setting to the configuration file:
Extra configuration settings for server ... WARNING: Typically, additional settings should only be added after consultation with Altair to address any specific requirements for Graph Lakehouse deployment in your environment or use for a specific application. []
Press Enter if you do not have a setting to add. If you have a custom
setting=valueto add, enter it in the prompt, and then press Enter.The installer asks if you want to start the system management service,
azgmgrd, after the installation is complete.Start azgmgrd service? Do you want to start azgmgrd systemd service? Yes [y], No [n, Enter]
Type y (Yes) and press Enter.
The installer asks if you want to start the database service,
anzograph, after the installation is complete.Start Graph Lakehouse service? Do you want to start Graph Lakehouse system services? Yes [y], No [n, Enter]
Type y (Yes) and press Enter.
The installer begins installing Graph Lakehouse based on your responses. When it is finished, the following message is displayed.
Setup has finished installing Graph Lakehouse on your computer. Finishing installation...
Now that Graph Lakehouse is installed and running, proceed to Complete the Post-Installation Configuration to complete the post-installation requirements.
Install Graph Lakehouse on a Cluster
Follow the steps below to install Graph Lakehouse on multiple servers in a cluster. There are two steps in the process:
Install Graph Lakehouse on the Compute Servers
Follow the instructions below to install Graph Lakehouse on each compute server.
Download the installer to each of the host servers. The installer is an interactive shell script that prompts you to choose configuration options for the deployment.
Change directories to the location where you copied the script and run the following command to make the script executable:
chmod +x <script_name>
For example:
chmod +x anzograph_linux_3_1_0_r202311010839.sh
Run the following command to start the installation wizard:
sudo ./<script_name>.sh
The installer first verifies that the prerequisites listed in Prerequisite Software are installed and presents a message if any are missing. For example, on a server where OpenJDK and OpenLDAP are installed but the glibc-devel package is missing, the installer displays the following message and the installation is canceled. In this case, the user would follow the instructions in GNU C Devel Library and then restart the installation.
Starting Installer ... Prerequisite software packages
The following packages are missing on your system and are required to be installed before proceeding with the Graph Lakehouse installation:
- glibc-devel
Canceling the installation, install the missing software, and then run the installer again.
If the prerequisite software is installed, but the C++ dependencies are not, the installer presents an informational message to let you know the dependencies are required but it is recommended that you install them after the Graph Lakehouse installation. For example:
Starting Installer ... Prerequisite software packages
INFO: The following C++ dependencies are also required and missing from your system. However, Altair recommends that you install them after the Graph Lakehouse installation is complete. Follow the Post-Installation instructions in the online documentation. libarmadillo12 libboost_filesystem1_80_0 libboost_iostreams1_80_0 libboost_system1_80_0 libgrpc++1 libflatbuffers2 libhdfs3 libnfs13 libserd-0-0 libsmb2
If all dependencies are installed or you proceed with the installation after getting the informational message, the license agreement is displayed. Press Enter to scroll through the text. At the end you are prompted to accept the agreement.
Starting Installer ... Please read the following License Agreement. You must accept the terms of this agreement before continuing with the installation.
ANZOGRAPH(R) DB END USER LICENSE AGREEMENT ... I accept the agreement Yes [1], No [2]
Enter 1 to accept the license agreement and continue the installation.
The installer prompts you to specify the directory where Graph Lakehouse should be installed.
Where should Graph Lakehouse be installed? [/opt/cambridgesemantics]
A number of subdirectories and an uninstall script are created inside the directory that you specify. One subdirectory is named
anzographand includes the installation files. Because an anzograph directory will be created, you may not want to specify/opt/anzographas the install location because that will result in an/opt/anzograph/anzographdirectory.Press Enter to select the default installation directory or specify an alternate location and then press Enter. The installation path must be the same on all servers in the cluster.
Next, the wizard prompts you to specify the installation type: single standalone server, leader server, or compute server:
Type of server being installed. Server Installation Type Standalone [1, Enter], Cluster Leader [2], Cluster Compute/Worker [3]
At the server installation type prompt, type 3 (Cluster Compute/Worker) and press Enter.
Next, the installer prompts you to specify the service user name. The default name is
anzograph.Setup Graph Lakehouse service user name Graph Lakehouse service user [anzograph]
Type the name of the Graph Studio service user. Typically it is anzo. Then press Enter.
Next, the wizard prompts you to specify the IP addresses for each of the servers in the cluster:
IP Address of nodes in cluster. Comma separated list of Cluster Nodes' IP Addresses. Leader node address is always first. Order must be the same on all nodes in cluster. [127.0.0.1]
Type a comma-separated list of the IP addresses for each server in the cluster. Type the leader server IP address first, followed by each compute IP address. For example, on a cluster with 4 servers where 192.168.2.1 is the leader server:
192.168.2.1,192.168.2.2,192.168.2.3,192.168.2.4
Make sure that you enter this value exactly the same, with IP addresses in the same order, during the installation on each server.
After listing the IP addresses for the cluster, press Enter.
The installer asks if you want to start the system management service,
azgmgrd. This service runs on all servers in the cluster and manages communication between the servers.Start azgmgrd service? Do you want to start azgmgrd systemd service? Yes [y], No [n, Enter]
Type y (Yes) and press Enter.
The installer begins installing Graph Lakehouse based on your responses. When it is finished, the following message is displayed.
Setup has finished installing Graph Lakehouse on your computer. Finishing installation...
Repeat the steps above to install Graph Lakehouse on each compute server. When the installation is complete and azgmgrd is running on all compute servers, proceed to Install Graph Lakehouse on the Leader Server below.
Install Graph Lakehouse on the Leader Server
Follow the instructions below to install Graph Lakehouse on the leader server.
Download the installer to the host server.
Change directories to the location where you copied the script and run the following command to make the script executable:
chmod +x <script_name>
For example:
chmod +x anzograph_linux_3_1_0_r202311010839.sh
Run the following command to start the installation wizard:
sudo ./<script_name>.sh
The installer first verifies that the prerequisites listed in Prerequisite Software are installed and presents a message if any are missing. For example, on a server where OpenJDK and OpenLDAP are installed but the glibc-devel package is missing, the installer displays the following message and the installation is canceled. In this case, the user would follow the instructions in GNU C Devel Library and then restart the installation.
Starting Installer ... Prerequisite software packages
The following packages are missing on your system and are required to be installed before proceeding with the Graph Lakehouse installation:
- glibc-devel
Canceling the installation, install the missing software, and then run the installer again.
If the prerequisite software is installed, but the C++ dependencies are not, the installer presents an informational message to let you know the dependencies are required but it is recommended that you install them after the Graph Lakehouse installation. For example:
Starting Installer ... Prerequisite software packages
INFO: The following C++ dependencies are also required and missing from your system. However, Altair recommends that you install them after the Graph Lakehouse installation is complete. Follow the Post-Installation instructions in the online documentation. libarmadillo12 libboost_filesystem1_80_0 libboost_iostreams1_80_0 libboost_system1_80_0 libgrpc++1 libflatbuffers2 libhdfs3 libnfs13 libserd-0-0 libsmb2
If all dependencies are installed or you proceed with the installation after getting the informational message, the license agreement is displayed. Press Enter to scroll through the text. At the end you are prompted to accept the agreement.
Starting Installer ... Please read the following License Agreement. You must accept the terms of this agreement before continuing with the installation.
ANZOGRAPH(R) DB END USER LICENSE AGREEMENT ... I accept the agreement Yes [1], No [2]
Enter 1 to accept the license agreement and continue the installation. The installer prompts you to specify the directory where Graph Lakehouse should be installed.
Where should Graph Lakehouse be installed? [/opt/cambridgesemantics]
Press Enter to select the default installation directory or specify an alternate location and then press Enter. The installation path must be the same on all servers in the cluster.
Next, the wizard prompts you to specify the installation type: single standalone server, leader server, or compute server:
Type of server being installed. Server Installation Type Standalone [1, Enter], Cluster Leader [2], Cluster Compute/Worker [3]
At the server installation type prompt, type 2 (Cluster Leader) and press Enter.
The next prompt asks you to create the username for the administrator. This username will be specified when creating the connection between Graph Studio and Graph Lakehouse.
Set up the Graph Lakehouse Admin user. Graph Lakehouse Admin user [admin]
Press Enter to accept the default username, admin, or type another username (case-sensitive) and press Enter. The installer asks you to create a password for the Admin user.
Graph Lakehouse Admin password
Enter the case-sensitive password for the Admin user and then press Enter. It is recommended that you create a non-trivial secure password that includes some combination of upper and lower case letters and digits. For security purposes, the installer does not echo the characters that you type.
There is no limitation placed on the length of the password. However, some special characters, such as
$and*, are treated as parameters in bash. When typing the password, avoid special characters. For more information, see Quoting in the Bash Reference Manual.Next, the installer asks if the installation will be used with Graph Studio:
Is this Graph Lakehouse installation intended for use with Graph Studio? Yes [y, Enter], No [n]
Press Enter for Yes. Answering yes configures Graph Lakehouse to use the settings that are optimal for Graph Studio. Answering no would configure the settings that are optimal for Graph Lakehouse standalone use without Graph Studio.
The installer asks if you want the admin user to be able to run queries.
Do you want to use admin user to run queries? Yes [y, Enter], No [n]
Press Enter for Yes. Answering yes is required for use with Graph Studio.
Next, the installer prompts you to specify the platform service user name. The default name is
anzo.Setup Graph Lakehouse service user name Graph Lakehouse service user [anzo]
Press Enter to accept the default name,
anzo, or type an alternate name and then press Enter if a different name is used for the account.Next, the wizard prompts you to specify the IP addresses for each of the servers in the cluster:
IP Address of nodes in cluster. Comma separated list of Cluster Nodes' IP Addresses. Leader node address is always first. Order must be the same on all nodes in cluster. [127.0.0.1]
Type a comma-separated list of the IP addresses for each server in the cluster. Type the leader server IP address first, followed by each compute IP address. For example, on a cluster with 4 servers where 192.168.2.1 is the leader server:
192.168.2.1,192.168.2.2,192.168.2.3,192.168.2.4
Make sure that you enter this value exactly the same, with IP addresses in the same order, during the installation on each server.
After specifying the IP addresses for the cluster, press Enter.
Next, the installer asks if you want to add a specific configuration setting to the configuration file:
Extra configuration settings for server ... WARNING: Typically, additional settings should only be added after consultation with Altair to address any specific requirements for Graph Lakehouse deployment in your environment or use for a specific application. []
Press Enter if you do not have a setting to add. If you have a custom
setting=valueto add, enter it in the prompt, and then press Enter.The installer asks if you want to start the system management service,
azgmgrd, after the installation is complete.Start azgmgrd service? Do you want to start azgmgrd systemd service? Yes [y], No [n, Enter]
Type y (Yes) and press Enter.
The installer asks if you want to start the database service,
anzograph, after the installation is complete.Start Graph Lakehouse service? Do you want to start Graph Lakehouse system services? Yes [y], No [n, Enter]
Type y (Yes) and press Enter.
The installer begins installing Graph Lakehouse based on your responses. When it is finished, the following message is displayed.
Setup has finished installing Graph Lakehouse on your computer. Finishing installation...
Now that Graph Lakehouse is installed and the cluster is running, proceed to Complete the Post-Installation Configuration to complete the post-installation requirements.
Source: https://docs.sw.siemens.com/documentation/external/PL20260212925461721/en-US/graph_studio/anzograph-installer.htm · retrieved 2026-08-23