Azure DevOps Services | Azure DevOps Server 2022 - Azure DevOps Server 2019
This article defines operational and object limits placed on work tracking operations and work tracking customization. In addition to the specified hard limits on specific objects, some practical limits apply. When you customize work item types (WITs), consider the limits placed on objects.
Work items and queries
When you define work items or running queries, keep the following operational limits in mind:
Object
Limit
Attachments added to a work item
100
Attachment size
60 MB
Long text field
1-M characters
Query execution time
30 seconds
Query results
20,000 items
Query length
32,000 characters
Shared queries under a folder
999 queries
Work item links assigned to a work item
1,000
Work item tags assigned to a work item
100
Work item revisions (REST API)
10,000
Favorite queries per project
200 queries
The REST API for Azure DevOps Services enforces a work item revision limit of 10,000 updates. This limit restricts updates made through the REST API, but updates from the web portal aren't affected.
500 per project. Accessible at the project level and anyone with access to the project can use.
Team Dashboards
500 per team. Specific to the team and used to track team-specific metrics and data.
Teams
5,000 per project
Work item tags
150,000 tag definitions per organization or collection
Delivery plans per project
1,000
Templates per work item type
100
Each backlog can display up to 10,000 work items. This limit applies to what the backlog can display, not to the number of work items you can define, as there's no specific limit. If your backlog exceeds this limit, consider adding a team and moving some work items to the new team's backlog.
Padoms
If you're approaching the dashboards limits, see the following steps to manage and clean up your dashboards:
Review usage: Identify dashboards that are no longer in use or are duplicates. You can do this by checking the last accessed date or by consulting with team members.
Consolidate dashboards: Combine similar dashboards to reduce the total number. This can be done by adding multiple widgets to a single dashboard.
Archive old dashboards: If certain dashboards are no longer needed but you want to keep the data, consider exporting the data and archiving the dashboards.
Completed or closed work items don't display on backlogs and boards once their Changed Date is older than a year. You can still list these items using a query. To make them show up on a backlog or board, make a minor change to reset the display clock.
By default, work item limits might be set to lower values initially.
When you work with teams, work item tags, backlogs, and boards, the following operational limits apply. Default and maximum limits.
User interface
Limit
Backlogs
999 work items
Boards
400 cards
Dashboards per project
500
Taskboard
800 work items
Teams
5,000 per project
Work item tags
150,000 tag definitions per project
Templates per work item type
100
Each backlog can display up to 999 work items. If your backlog exceeds this limit, consider creating a team and moving some of the work items to the new team's backlog.
For the On-premises XML process model, you can modify the backlog and Taskboard limits by editing the ProcessConfiguration.xml file. For details, see Process configuration XML element reference.
Azure DevOps Services limits each organization to 1,000 projects per organization, an increase over the previous limit of 300 projects.
Piezīme
Above 300 projects, certain experiences, like connecting to a project from Visual Studio, might degrade. For on-premises Azure DevOps Server, there are no hard limits, but performance issues may arise as the number of projects nears 300. When migrating to Azure DevOps Services, observe the maximum limit of 1,000 projects. If your collection exceeds this limit, split the collection or delete older projects.
The following table lists the maximum number of objects you can define for the Inheritance and Hosted XML process models. While these limits are hard limits, practical limits might also apply.
Object
Inheritance
Hosted XML
Number of processes you can have in an organization
128
64
Work item types defined for a process
64
64
Fields defined for an organization
8192
8192
Fields defined for a process
1024
1024
Fields defined for a work item type
1024
1024
Picklists defined for an organization or collection
For the Hosted XML process model, you can define approximately 10,000 items across all global lists specified in all WITs.
The following table lists the maximum number of objects you can define for the Inheritance and On-premises XML process models. While these limits are hard limits, practical limits might also apply.
Object
Inheritance
On-premises XML
Number of processes you can have in an organization
64
64
Work item types defined for a process
64
64
Fields defined for a collection
8192
1024
Fields defined for a process
1024
1024
Fields defined for a work item type
1024
1024
Picklists defined for a collection
1024
N/A
Picklist items defined for a list
2048
2048
Picklist item character length
256
N/A
Workflow states defined for a work item type
32
16
Rules defined for a work item type
1024
1024
Portfolio backlog levels defined for a process
5
5
Categories defined for a process
N/A
32
Global lists defined for a process
N/A
256
List items defined within a global list
N/A
1024
Piezīme
For the On-premises XML process model, you can define an approximate total of 10K items for all global lists specified across all WITs.
Practical limits
To minimize performance issues, we recommend following this guidance:
Limit the number of custom fields you define. All custom fields contribute to the total allowed for a process, collection, or organization. You can specify different behaviors, such as rules and picklists, for the same field in different WITs.
Limit the number of rules you define for a WIT. While you can create multiple rules for a WIT, other rules can negatively affect performance when users add or modify work items. When users save work items, the system validates all rules associated with the fields for that work item type. In some cases, the rule validation expression might be too complex for SQL to evaluate efficiently.
Limit the number of custom WITs you define.
Limit the number of custom fields you define. All custom fields contribute to the total allowed for a process, collection, or organization. You can specify different behaviors, such as rules and picklists, for the same field in different WITs.
Limit the number of rules you define for a WIT. While you can create multiple rules for a WIT, other rules can negatively affect performance when users add or modify work items. When users save work items, the system validates all rules associated with the fields for that work item type. In some cases, the rule validation expression might be too complex for SQL to evaluate efficiently.
Limit the number of custom WITs you define.
Limit the number of reportable fields you define. Reportable fields can affect the performance of your data warehouse.
Piezīme
Work Item Rules Validation Exceeds SQL Limits: A single SQL expression is defined per project to validate work items whenever they are created or updated. This expression grows with the number of rules specified for all work item types in the project. Each behavioral qualifier for a field increases the number of sub-expressions. Nested rules, rules that apply only on a transition, or rules conditioned on the value of another field add more conditions to an IF statement. Once the expression reaches a certain size or complexity, SQL can no longer evaluate it and generates an error. To resolve this error, remove some WITs or eliminate some rules.
Rate limits
To reduce costs and enhance scalability and performance, Azure DevOps Services, like many Software-as-a-Service solutions, uses multi-tenancy. To ensure good performance and minimize the risk of outages, Azure DevOps Services limits the resources individuals can consume and the number of requests they can make to certain commands. When these limits are exceeded, subsequent requests might be delayed or blocked.
Pievienojieties meetup sērijai, lai kopā ar citiem izstrādātājiem un ekspertiem izveidotu mērogojamus AI risinājumus, kuru pamatā ir reālas lietošanas gadījumi.