Admin Menus

When you log in with an admin account, you will see an additional Administration menu in the bottom left sidebar. User information registered in Backend.AI is listed in the Users tab. You can view active users in the Active tab and inactive users in the Inactive tab. Admins can see all user information. Only admins can create and deactivate a user.

User ID (email) and Name (username) can be filtered by typing some text in the search box on each column header.

User management page

Create and update user

A user can be created by clicking the CREATE USER button. Note that the password must be longer or equal to 8 characters and at least 1 alphabet/special character/ number should be included. The maximum allowed length of E-Mail and Username is 64.

Note

If user with the same email or username already exists, it is not possible to creat user account. Please try other email and username.

../_images/create_user_dialog.png

If password didn’t match the policy, then validation failed message will be displayed under the input field. A message displays the reason why the password you entered is wrong, so you can easily check it.

  • no inputs
../_images/create_user_dialog_wrong.png
  • no input on password confirmation
../_images/create_user_dialog_wrong2.png
  • all input with non-matching value
../_images/create_user_dialog_wrong3.png

Check if the user is created.

User management page

Click the green button in the Controls column for more detailed user information. You can also check the domain and group information where the user belongs.

Detailed information of a user

Click the gear icon in the Controls column to update information of a user who already exists. User’s name, password, activation state, etc. can be changed. Notice that user ID is disabled so that it can’t be changed.

Also, password can be updated with same policy as creating a new user. Username and Fullname can contain up to 64 characters.

User update dialog

Each of the two items at the bottom of this dialog has the following functions.

  • Active user?: Indicates the user’s active status. Inactive users cannot log in. You can adjust this option to change the user to active or inactive state. Note that for inactive users, they are displayed separately in the Inactive tab.
  • Require password change?: If an admin has randomly specified a password while creating users in bulk, this field can be set to On to distinguish between them. This is a kind of descriptive flag indicating that password change is required, but it has no effect on actual use.

Deactivate user account

Deleting user accounts is not allowed even for superadmins, to track usage statistics per user, metric retention, and accidental account loss. Instead, admins can deactivate user accounts to keep users from logging in. Click the trash icon in the Controls column. A dialog asking confirmation appears, and you can deactivate the user by clicking the OKAY button.

Deactivating user account

To re-activate users, go to Users - Inactive tab, and edit the target user to turn on “Active user?” field.

Note

Please note that inactivating the user affects all of credentials to be inactive, but reactivating the user does not affects inactivated credentials, since the user account can have multiple keypairs, which brings hard to decide which credential should be reactivated.

Sharing a group storage folder with group members

In addition to user’s storage folders, Backend.AI provides storage folders for groups. A group storage folder is a folder belonging to a specific group, not a specific user, and can be accessed by all users belonging to that group.

Note

Group folders can only be created by administrators. Normal users can only access the contents of the group folder created by the administrator. Also, there may be cases where group folder is not allowed depending on the system settings.

First, log in with an administrator account and create a group folder. After moving to the Data & Storage page, click NEW FOLDER to open the folder creation dialog. Enter the folder name, set the Type to Group, and select the target group. The target group should be set to the group to which User B belongs for this example to work. Permission is set to Read-Only again.

../_images/group_folder_creation.png

After confirming that the folder has been created, log in with the user B’s account and check that the group folder just created on the Data & Storage page is displayed without an invitation procedure! You can see that R (Read Only) is also displayed in the Permission column.

../_images/group_folder_listed_in_B.png

In this way, Backend.AI provides invitation for user’s folders and group folders to help you share data and models in a flexible way.

Limiting Maximum Compute Resources per User (Keypair)

Note

Objective

  • After logging in to the administrator account, use the function to set the allowed resources for each user (keypair)
  • Confirm that only allowed resources can be used at maximum when creating a compute session by a user

In Backend.AI, you can set a limit on the total amount of resources available for each user and group. Resource limits per group can only be set through the enterprise-dedicated administrator GUI Hub, but resource limits per user (precisely user’s keypair) can be set in the user GUI console.

Log in to the user GUI Console with an administrator account to check the resource limit status. If you log in as an administrator, you will see more administrator-only menus at the bottom of the left menu. Among them, click Users to go to the user management page and click the Resource Policies tab. You can check the list of currently registered resource policies. In the figure below, there are a total of three policies (gardener, STUDENT, default). If you look at the Resources column of the default policy, you can see that resources such as Cores, Ram, and fGPU are marked as infinite (∞). The infinity means that no resource restrictions have been placed at all.

../_images/resource_policy_page.png

Currently, the user accounts used in this guide are restricted by the default resource policy. This can be found in the Credentials tab on the left. If you look at the Resource Policy, you can also see that all resources can be used indefinitely (to the extent the hardware allows).

../_images/credentials.png

Log in again with your user account, now. Go to the Sessions page and check the resource indicator at the top. Check the maximum quota for CPU, RAM and FGPU. This corresponds to the total amount of resources currently held by the test node.

../_images/resource_indicator_no_limit.png

Click START to open the compute session creation dialog and pull the CPU, RAM, and GPU sliders all the way to the right, and you can see that you cannot allocate more than the total amount shown in the resource indicator.

../_images/session_launch_dialog_no_limit.png

Now log back into the administrator account and go to the Resource Policies tab. Click the gear button in the Control column of the default policy to enter the policy editing screen. In Update Resource Policy dialog, every option is changeable except Policy Name which is the primary key for distinguishing the resource policy from list. Each numeric value is regarded as maximum value for creating/destroying a session and also for the number of vfolders. Uncheck the Unlimited checkbox at the bottom of CPU, RAM, fGPU, and set the resource limits to the desired values. Set the amount of constrained resource to be smaller than the total amount of hardware. Here, we set it to 2, 4 and 1 respectively. Click the UPDATE button to update the resource policy.

../_images/update_resource_policy.png

About details in each option in resource policy dialog, see the description below.

  • Resource Policy
    • CPU : Specify the maximum amount of CPU cores. (max value: 512)
    • RAM : Specify the maximum amount of memory in GB. It would be good practice to set memory twice as large as the maximum value of GPU memory. (max value: 1024)
    • GPU : Specify the maximum amount of physical GPUs. If fractional GPU is enabled by the server (it is in most Enterprise sites), this setting has no effect. (max value: 64)
    • fGPU : Fractional GPU (fGPU) is literally split a single GPU to multiple partitions in order to use GPU efficiently. Notice that the minimum amount of fGPU required is differed by each image. If fractional GPU is not enabled by the server, this settings has no effect. (max value: 256)
  • Sessions
    • Container Per Session : The maximum number of containers per session. Currently, this value has no effect since the server only allows one container per compute session. (max value: 100)
    • Idle timeout (sec.) : Configurable period of time during which the user can be inactive without any impact on their session. If there is no activity at all on a compute session for idle timeout, the session will be garbage collected and destroyed automatically.(max value: 15552000 (approx. 180 days))
    • Concurrent Jobs : Maximum number of concurrent compute session per keypair. If this value is set to 3, for example, user bound to this resource policy caanot create more than 3 compute sessions simultaneously. (max value: 100)
  • Folders
    • Allowed hosts : Backend.AI supports many NFS mountpoint. This field limits the accessibility to them. Even if a NFS named “data-1” is mounted on Backend.AI, user cannot access it unless it is allowed by resource policy.
    • Capacity(GB) : the maximum size (GB) a storage folder can contain. This feature is only effective for special type of storages/filesystems such as FlashBlade. (max value: 1024)
    • Max. # : the maximum number of storage folders that can be created/invited. (max value: 50)

In the resource policy list, check that the Resources value of the default policy has been updated.

../_images/update_check.png

Once again, log in with your user account. Go to the Session page. You can see that the resource constraint value set earlier is reflected in the resource indicator at the top.

../_images/resource_indicator_limit.png

Click START to open the calculation session creation dialog and pull the CPU, RAM, and GPU sliders all the way to the right, and you can see that it cannot be moved beyond the value after the resource constraint was placed.

../_images/session_launch_dialog_limit.png

In this way, it was confirmed that the development user can control the maximum amount of resources that can be allocated when creating a compute session based on the user’s key pair.

You can also delete each of resource keypairs by clicking trash can icon in the Control panel. When you click the icon, the confirmation dialog will appears. click OKAY button to delete.

../_images/resource_policy_delete_dialog.png

Note

If there’s any user including inactive user follows the resource policy to be deleted, It will throw an error. Please make sure that Before deleting the task, make sure that no users who have selected that resource policy remain.

Note

In each of USERS, CREDENTIALS, RESOURCE POLICIES tabs, there is an icon (...) on the far right side of the tab header. Clicking this reveals export CSV menu, which again brings up a CSV export dialog for each tabs. By giving appropriate file name, if necessary, and clicking EXPORT CSV FILE button, you can download the list of users, keypairs, and/or resource policies.

../_images/export_csv_user.png

Session List Download

There’s additional feature in Session page for admin. On the right side of the OTHERS tab there is a menu marked with .... When you click this menu, a sub-menu export CSV appears.

../_images/export_csv_menu.png

If you click this menu, you can download the information of the calculation sessions created so far in CSV format. After the following dialog opens, enter an appropriate file name (if necessary) and click the EXPORT button. Please note that file name can only contain up to 255 characters. A CSV file will be downloaded soon.

../_images/export_session_dialog.png

Server management

Go to the Maintenance page and you will see some buttons to manage the server.

  • RECALCULATE USAGE: Occasionally, due to unstable network connections or container management problem of Docker daemon, there may be a case where the resource occupied by Backend.AI does not match the resource actually used by the container. In this case, click the RECALCULATE USAGE button to manually correct the resource occupancy.
Maintenance page

Information

Information page display various details about Backend.AI service, license information.