title: Best Practices for Forefront Identity Manager (FIM) 2010 TOCTitle: Best Practices for FIM 2010 ms:assetid: 796d4a1a-4203-430b-b989-eca97cdbd184 ms:mtpsurl: https://technet.microsoft.com/en-us/library/Ff608274(v=WS.10) ms:contentKeyID: 30278712 ms.date: 05/28/2012 mtps_version: v=WS.10
Best Practices
Applies To: Forefront Identity Manager 2010
This topic describes the best practices for deploying and operating Microsoft® Forefront® Identity Manager (FIM) 2010.
Note
The following recommendations for setting up a server running SQL presume a SQL instance dedicated to the FIMService and to the FIMSynchronizationService databases. If you are running the FIMService in a consolidated environment, you will have to make adjustments appropriate for your configuration.
Configuration of the Structured Query Language (SQL) server is critical to optimal system performance. Achieving optimum FIM performance in large-scale implementations depends on the application of best practices for a server running SQL. For more information, see the following topics about SQL best practices:
Storage Top 10 Best Practices (https://go.microsoft.com/fwlink/?LinkID=183663)
Optimizing tempdb Performance (https://go.microsoft.com/fwlink/?LinkID=188267)
SQL Server Best Practices Article (https://go.microsoft.com/fwlink/?LinkID=188268)
Reorganizing and Rebuilding Indexes (https://go.microsoft.com/fwlink/?LinkID=188269)
Do not rely on autogrow. Instead, manage the growth of these files manually. You can leave autogrow on for safety reasons, but you should proactively manage the growth of the data files. For sample sizes of the FIM database, see the FIM Capacity Planning Guide (https://go.microsoft.com/fwlink/?LinkID=185246).
Start SQL Server Enterprise Manager.
Navigate to the database FIMService, right-click FIMService, and then click Properties.
On the Files page, expand the database files to the required size.
Follow the SQL server best practices to isolate the transaction and data log files for the databases onto separate physical disks.
For optimal performance, we recommend that you create one data file per CPU core in the tempdb file.
Start SQL Server Enterprise Manager.
Navigate to the database tempdb in System Databases, right-click tempdb, and then click Properties.
On the Files page, create one data file for each CPU core. Be sure to separate the tempdb data and log files to different drives and spindles.
It is important to understand your recovery model’s disk requirements. Simple recovery mode may be appropriate during the initial system load to limit the use of your disk space, but the data created after your most recent backup is exposed to data loss. When using Full recovery mode, you need to manage the disk usage through backups which include frequent backups of the transaction log to prevent high disk space usage. For more information, see Recovery Model Overview (https://go.microsoft.com/fwlink/?LinkID=185370).
Depending on how much memory you have on your SQL server and if you share the SQL server with other services (that is, FIM 2010 R2 Service and FIM 2010 R2 Synchronization Service), you might want to restrict the memory consumption of SQL. You can do this through the following steps.
Start SQL Server Enterprise Manager.
Select New Query.
Run the following query:
USE master EXEC sp_configure 'show advanced options', 1 RECONFIGURE WITH OVERRIDE USE master EXEC sp_configure 'max server memory (MB)', 12000--- max=12G RECONFIGURE WITH OVERRIDE
This example reconfigures the SQL server to use no more than 12 gigabyte (GB) of memory.
Verify the setting by using the following query:
USE master EXEC sp_configure 'max server memory (MB)'--- verify the setting USE master EXEC sp_configure 'show advanced options', 0 RECONFIGURE WITH OVERRIDE
In general, you should perform database backups according to your organization’s backup policy. If incremental log backups are not planned, the database should be set to the Simple recovery mode. Ensure that you understand the implications of the different recovery models before implementing your backup strategy as well as the disk space requirements for these models. Full recovery model requires frequent log backups to avoid high disk space usage. For more information, see Recovery Model Overview (https://go.microsoft.com/fwlink/?LinkID=185370) and FIM 2010 Backup and Restore Guide (https://go.microsoft.com/fwlink/?LinkID=165864).
Important
Members of the FIMService Administrators set have unique permissions critical to the operation of your FIM deployment. If you are unable to logon as part of the Administrators set, the only resolution is to roll back to a previous backup of the system. To mitigate this situation, we recommend that you add other users to the FIM Administrative set as part of your post-installation configuration.
The following are best practices for configuring Microsoft Exchange Server for the FIM 2010 R2 Service service account.
Configure the service account so that it can accept mail only from internal e-mail addresses. Specifically, the service account mailbox should never be able to receive mail from external SMTP servers.
To configure the service account
In the Exchange Management Console, select the FIM Service service account.
Select Properties, select Mail Flow Settings, and then select Mail Delivery Restrictions.
Select the Require that all senders are authenticated check box.
For further information, see Configure Message Delivery Restrictions (https://go.microsoft.com/fwlink/?LinkID=183625).
Configure the service account so that it rejects mail with sizes greater than 1 MB. Follow the best practice to Configure Message Size Limits for a Mailbox or a Mail-Enabled Public Folder (https://go.microsoft.com/fwlink/?LinkID=183626).
Configure the service account so that it has a mailbox storage quota of 5 GB. For optimum results, follow the best practices listed in Configure Storage Quotas for a Mailbox (https://go.microsoft.com/fwlink/?LinkID=156929).
We recommend that you disable Microsoft Office SharePoint® indexing. There are no documents that need to be indexed, and indexing causes many error log entries and potential performance problems with FIM 2010.
On the server that hosts the FIM 2010 Portal, click Start.
Click All Programs.
In the All Programs list, click Administrative Tools.
Under Administrative Tools, click SharePoint 3.0 Central Administration.
On the Central Administration page, click Operations.
On the Operations page, under Global Configuration, click Timer job definitions.
On the Timer Job Definitions page, click SharePoint Services Search Refresh.
On the Edit Timer Job page, click Disable.
This section lists a series of steps to increase the performance of the initial data load from external system to FIM 2010. It is important to understand that a number of these steps are temporary during the initial population of the system and should be reset upon its completion. This is a one-time operation and is not a continuous synchronization.
Note
For more information about synchronizing users between FIM 2010 and Active Directory Domain Services (AD DS), see How do I Synchronize Users from Active Directory to FIM (https://go.microsoft.com/fwlink/?LinkID=188277) in the FIM documentation.
Important
Ensure that you have applied the best practices covered in the SQL setup section of this guide.
When you plan to initially load a lot of data, you can shorten the time it takes to populate the database by temporarily turning off the full-text search and turning it on again after the export on the FIM 2010 R2 management agent (FIM MA) has completed.
Start SQL Enterprise Manager.
Select New Query.
Run the following SQL statements:
ALTER FULLTEXT INDEX ON [fim].[ObjectValueString] SET CHANGE_TRACKING = MANUAL
ALTER FULLTEXT INDEX ON [fim].[ObjectValueXml] SET CHANGE_TRACKING = MANUAL
It is important that you understand the disk requirements for your SQL server’s recovery model. Depending on your backup schedule, you may consider using the Simple recovery mode during the initial system load to limit the disk space usage, but you need to understand the implications from a data loss perspective. When using Full recovery mode, you need to manage the disk usage through backups which includes frequent backups of the transaction log to prevent high disk space usage.
Important
Not implementing these procedures can result in high disk space usage, possibly causing you to run out of disk space. You can find additional details about this topic in Recovery Model Overview (https://go.microsoft.com/fwlink/?LinkID=185370). The FIM Backup and Restore Guide (https://go.microsoft.com/fwlink/?LinkID=165864) contains additional information.
During the initial load process, you should only apply the minimum configuration required to your FIM configuration for your management policy rules (MPRs) and set definitions. After the data load is completed, create the additional sets required for your deployment. Use the Run-On Policy update setting on the action workflows to apply those policies retroactively on the loaded data.
At this point you should follow the procedures described in the How Do I Synchronize Users from Active Directory Domain Services to FIM guide to configure and synchronize your system with users from Active Directory. If you need to synchronize Group information, the procedures for that process are described in the How Do I Synchronize Groups from Active Directory Domain Services to FIM guide.
To optimize performance, run an export after a synchronization run that results in a large number of pending export operations in a connector space.
Then, run a confirming import run on the management agent that is associated with the affected connector space. For example, when you need to run synchronization run profiles on several management agents as part of an initial data load, you should run an export followed by a delta import after each individual synchronization run.
For each source management agent that is part of your initialization cycle, perform the following steps:
Full import on a source management agent.
Full synchronization on the source management agent.
Export on all affected target management agents with staged export operations.
Delta import on all affected target management agents with staged export operations.
Once your initial data load is completed, you should apply the full FIM configuration for your deployment. Depending on your scenarios, this may include the creation of additional sets, MPRs, and workflows. For any policies that you need to apply retroactively to all existing objects in the system, use the run-on policy update setting on action workflows to apply those policies retroactively on the loaded data.
Remember to change the SQL setting to its normal settings. This includes:
Turning on the full-text search
Updating your backup policy per your organization policy
Once you have completed the initial data load, you need to turn on full-text search again. Run the following SQL statements to turn on full-text search again:
ALTER FULLTEXT INDEX ON [fim].[ObjectValueString] SET CHANGE_TRACKING = AUTO
ALTER FULLTEXT INDEX ON [fim].[ObjectValueXml] SET CHANGE_TRACKING = AUTO
If you have to switch to Simple recovery mode, ensure that you reconfigure your backup schedule in accordance with your organization’s backup policy. Additional details of FIM backup schedules are available in the FIM Backup and Restore Guide (https://go.microsoft.com/fwlink/?LinkID=165864).
For many object types such as MPRs, the syncproduction.ps1 script uses the display name as the only anchor attribute between two systems. Consequently, a change to an existing MPR’s display name results in the deletion of the existing MPR, followed by the creation of a new MPR. This result occurs because the migration process cannot successfully join MPRs whose join criteria have changed. To avoid this issue, you can bind a custom attribute to all configuration object types and use that attribute as the join criteria. This enables you to modify display names without affecting the migration process.
While the file format and application programming interface (API) of the low-level objects are public and manipulations are supported by developers, we do not recommend that you change the contents of the intermediate formats during the migration. However, it may be necessary to remove entire ImportObjects from changes.xml or to perform find and replace operations on pilot.xml to replace version numbers or pilot Domain Name System (DNS) information for production DNS information.
While migrations across version numbers are not recommended or supported, you can often do this by replacing the pilot version number with the production version number in pilot.xml. Specifically, WorkflowDefinition and ActivityInformationConfiguration objects require the version number to refer precisely to workflow activities in the production environment. Failing to replace the version number results in the Compare-FIMConfig cmdlet identifying differences between the Extensible Object Markup Language (XOML) attributes on WorkflowDefinitions and migrating the pilot’s version number. The production FIM Service may fail to start workflow activities with the incorrect version number.
In general, cyclic references are not recommended in a FIM configuration. However, cycles sometimes occur when Set A refers to Set B and Set B also refers to Set A. To avoid issues with cyclic references, you should change the definition of Set A or Set B so that they both do not refer to each other. Then, restart the migration process. If you do have cyclic references and the Compare-FIMConfig cmdlet results in an error as a result, it is necessary to break the cycle manually. Because the Compare-FIMConfig cmdlet outputs a list of changes in order of precedence, it requires that no cycles exist among the references of configuration objects.
The FIM MA account is not considered a service account and should be a regular user account. The accounts must be able to log on locally in order for the FIM Synchronization Service service account to be able to impersonate it.
Click Start, click Administrative Tools, and then click Local Security Policy.
Open the Local Policies node, and then click User Rights Assignment.
In the policy Allow log on locally, ensure that the FIM MA account is explicitly specified, or add it to one of the groups that is already granted access.
To configure the servers running the FIM server components in a secure manner, the service accounts should be restricted. Using the previous procedure to turn on the FIM MA account, set the following restrictions on the FIM Synchronization Service and FIM Service accounts:
Deny logon as a batch job
Deny logon locally
Deny access to this computer from the network
The service accounts should not be a member of the local administrators group.
The FIM Synchronization Service service account should not be a member of the security groups used to control access to FIM Synchronization Service (groups starting with FIMSync, for example, FIMSyncAdmins, and so on).
Important
If you select the options to use the same account for both service accounts and you separate the FIM Service and the FIM Synchronization Service, then you cannot set Deny access to this computer from the network on the mms Synchronization Service server. If access is denied that would prohibit the FIM Service from contacting the FIM Synchronization Service to change configuration and manage passwords.
Password reset deployed to kiosk-like computers should set local security to clear virtual memory pagefile
When deploying FIM password reset on a workstation intended to be a kiosk, we recommend that the Shutdown: Clear virtual memory pagefile local security policy setting be turned on to ensure that sensitive information from process memory is not available to unauthorized users.
It is highly recommended that you use secure sockets layer (SSL) on the FIM Portal server to secure the traffic between the clients and the server.
On the FIM Portal server, open IIS Manager.
Click the local computer name.
Click Server Certificates.
Click Create Certificate Request.
In the Common Name text box, enter the name of the server.
Click Next, and then click Next.
Save the file to any location. You will need to access this location in subsequent steps.
In Windows Internet Explorer®, browse to https://servername/certsrv. Replace servername with the name of the server issuing certificates.
Click Request a new Certificate.
Click Submit an Advanced Request.
Click Submit a Certificate Request by using a base-64-encoded.
Paste the contents of the file that you saved in the previous step.
In Certificate Template, select Web Server.
Click Submit.
Save the certificate to your desktop.
In IIS Manager, click Complete Certification Request.
Point IIS Manager to the certificate you just saved to the desktop.
For Friendly name, type the name of the server.
Click Sites, and then select SharePoint – 80.
Click Bindings, and then click Add.
Select https.
For certificate, select the one that has the same name as the server (this is the certificate that you just imported).
Click OK.
Remove the HTTP binding.
Click SSL Settings, and then check Require SSL.
Save the settings.
Click Start, click Administrative Tools, and then click SharePoint 3.0 Central Administration.
Click Operations, and then click Alternate Access Mappings.
Click https://servername.
Change https://servername to https://servername, and then click OK.
Click Start, click Run, type iisreset, and then click OK.
For optimal performance configuration:
Apply the SQL setup best practices as described in the SQL setup section in this document.
Turn off SharePoint Indexing on the FIM 2010 R2 Portal site. For more information, see the Disable SharePoint indexing section in this document.
By default FIM 2010 purges expired system objects, which includes completed requests with associated approvals, approval responses, and workflow instances on a 30-day interval. If your organization needs a longer request history, you should export requests from FIM and store them in an auxiliary database to preserve them beyond the 30-day window. While the 30-day request deletion window is configurable, extending this window can negatively impact performance due to the additional objects in the system.
FIM provides two types of MPRs, Request and Set Transition:
Request MPR (RMPR)
Used to define the access control policy (authentication, authorization, and action) for Create, Read, Update, or Delete (CRUD) operations against resources.
Applied when a CRUD operation is issued against a target resource in FIM.
- Scoped by the matching criteria defined in the rule, that is, to which CRUD requests the rule applies.
Set Transition MPR (TMPR)
Use to define policies regardless of how the object entered the current state represented by the Transition Set. Use TMPR to model entitlement policies.
Applied when a resource enters or leaves an associated set.
- Scoped to the members of the set.
Note
For additional details, see Designing Business Policy Rules (https://go.microsoft.com/fwlink/?LinkID=183691).
Use the principle of least privilege when applying your configuration. MPRs control the access policy to your FIM deployment. Enable only those features used by most of your users. For example, not all users use FIM for group management, so associated group management MPRs should be disabled. By default, FIM ships with most nonadministrator permissions disabled.
When needing to modify the built-in MPRs, you should create a new MPR with the required configuration and turn off the built-in MPR. This ensures that any future changes to the built-in MPRs that are introduced through the upgrade process do not negatively impact your system configuration.
Using explicit attribute lists helps to prevent the accidental granting of permissions to nonprivileged users when attributes are added to objects. Administrators should explicitly need to grant access to new attributes instead of trying to remove access.
Access to data should be scoped to the business needs of the users. For example, group members should not have access to the filter attribute of the group they are a member of. The filter can inadvertently reveal organizational data that the user would not normally have access to.
Avoid granting permissions to attributes that the user can never use. For example, you should not grant permission to modify core resource attributes such as objectType. Despite the MPR, any attempt to modify a resource's type after it is created is denied by the system.
Read permissions should be separate from Modify and Create permissions when using explicit attributes in MPRs
When explicitly listing attributes in MPRs, the attributes required for Create and Modify are usually different than the ones available for Read. For example, Read can be granted over System attributes such as Creator or objectId, while Create or Modify cannot be specified for System attributes.
Create permissions should be separate from Modify permissions when using explicit attributes in rules
The Create operation requires that the user select the objectType as part of its operation. This is a core system attribute that cannot be modified after a Create operation.
For attributes with the same access requirements that are not expected to change, you can combine them into a single request MPR for efficiency.
In FIM, permissions are defined as a positive assertion. Because FIM does not support Deny permissions, giving unrestricted access to a resource complicates providing any exclusions in the permissions. As a best practice, grant only the permissions necessary.
Use Set Transition MPRs (TMPRs) instead of RMPRs to define custom entitlements. TMPRs provide a state-based model to assign or remove entitlements based on the membership in the defined Transition Sets, or roles, and the accompanying workflow activities. TMPRs should always be defined in pairs, one for resources transitioning in, and one for resources transitioning out. In addition, each transition MPR should contain separate workflows for provisioning and deprovisioning activities.
Note
Any deprovisioning workflow should ensure that the Run On Policy Update attribute is set to true.
When creating a TMPR pair, turn on Set Transition In MPR last. This order ensures that no resource is left with the entitlement if it is added to and removed from the set while In MPR is turned on but before Out MPR is turned on.
Provisioning workflows should first check to determine if the target resource has already been provisioned in accordance with the entitlement. If it has, then it should do nothing.
Deprovisioning workflows should first check to determine if the target resource has been provisioned. If it has, then it should deprovision the target resource. Otherwise, it should do nothing.
This ensures that the correct provisioning behavior applies when policy updates are implemented and use the Run-On Policy update flag on action workflows associated with the TMPRs. This ensures that changes in the policy definitions apply the action workflows to new members of the Transition Set.
Associating the same entitlement with two different Transition Sets can cause an unnecessary revoking and regranting of entitlements if the resource moves from one set to the other. As a best practice, ensure that one set contains all resources that require the associated entitlement. This ensures a one-to-one relationship between the Transition Set and the entitlement granting the workflow.
The order of the steps performed when removing entitlements in the system can result in two different operational results. Ensure that you understand which order applies to the effect that you want.
To remove an entitlement from the system (and revoke it from all members currently holding the entitlement):
Disable the T-In MPR. This avoids new grants.
Delete the T-Set filter or change it so that the set is empty. This causes all existing members to transition out and applies the transition out policy, including the configured deprovision workflows associated with the entitlement.
Disable the T-Out MPR.
To remove an entitlement but leave the current members alone (for example, stop using FIM to manage the entitlement):
Disable the T-In MPR. This avoids new grants.
Disable the T-Out MPR.
Delete the T-Set filter or change it so that the set is empty. Because the set is no longer tied to a TMPR, no deprovision workflows are applied.
When applying the best practices for sets, you need to consider the impact of the optimizations on the manageability and ease of future administration. Appropriate testing at expected production scale should be performed to identify the right balance between performance and manageability before applying these recommendations.
Note
All the following guidelines apply to dynamic sets and dynamic groups.
This refers to the filter of a set referencing the ComputedMember attribute of another set. A common reason for nesting sets is to avoid duplicating a membership condition across many sets. While this approach may result in better manageability of the sets, there is a performance tradeoff. You can optimize for performance by duplicating the membership conditions of a nested set instead of nesting the set itself.
You may encounter cases where you cannot avoid nesting sets to satisfy a functional requirement. These are the primary situations where you should nest sets. For example, to define the set of all the groups without Full-Time Employee owners, the nesting of sets must be used as follows: /Group[not(Owner = /Set[ObjectID = ‘X’]/ComputedMember], where ‘X’ is the ObjectID of the set of All Full Time Employees.
Negative conditions are the membership conditions that make use of the following operators or functions: !=, not(), < , <=. To optimize for performance, where possible, express the condition that you want with multiple positive conditions rather than as a negative condition.
The use of conditions based on multivalued reference attributes should be minimized because large numbers of these sets may affect the performance of operations on the attribute used in the membership condition.
Kiosk-like computers that are used for password reset should set local security to clear the virtual memory pagefile
When deploying FIM 2010 password reset on a workstation intended to be a kiosk, we recommend that the Shutdown: Clear virtual memory pagefile local security policy setting be turned on to ensure that sensitive information from the process memory is not available to unauthorized users.
When a user attempts to register for password reset through a Web portal, FIM 2010 always initiates registration on behalf of the logged-on user, regardless of who is logged onto the Web site. Users should always register for a password reset on a computer that they are logged on to.
When using FIM 2010 password reset, do not set the AvoidPdcOnWan registry key to true.
If this registry key is set to true, the user very likely goes through the password gates, has their password reset on the primary domain controller (PDC), and attempts to log on. Because of this registry key, the local domain controller does not perform the secondary validation with the PDC, therefore denying the logon request. If the user is denied enough times, they can be locked out of the domain and will need to call support.
It is possible to log clear-text passwords when turning on diagnostic Service Level tracing in Windows Communication Foundation (WCF). This option is not turned on by default, and you are discouraged from turning it on in production environments. These passwords are visible as clear-text elements within an encrypted Simple Object Access Protocol (SOAP) message when users register for password reset. For more information, see Configuring Message Logging (https://go.microsoft.com/fwlink/?LinkID=168572).
You should not attach an authorization workflow to a password reset operation. Password reset requires a synchronous response and authorization workflows that contain activities such as the approval activity are asynchronous.
You should not attach a workflow that contains more than one action activity to a password reset operation. An example scenario would be attaching a second AD DS password reset activity to a password reset MPR. This scenario is not supported.
Require reregistration when adding, removing, or changing the order of activities in an existing workflow
When adding, removing, or changing the order of authentication activities in an existing workflow, always select the option to require reregistration. Users who attempt to authenticate for password reset after an activity has been added to or removed from a workflow, but before they have reregistered, may encounter unwanted effects.
In FIM, by default, some user profile information may be displayed to other users. As a courtesy to the users, administrators should consider adding custom text consistent with their company's policies to the User Profile page. For more information about adding custom text to a FIM Portal page, see Introduction to Configuring and Customizing the FIM Portal (https://go.microsoft.com/fwlink/?LinkID=165848).
Though the Person and Group resource types are not marked as Core resource types, the resources themselves or the attributes assigned to them should not be deleted. The user interface (UI) in the FIM Portal requires that the Person and Group resource types and their attributes are present.
There are 13 Core attributes assigned to all resource types. You should not in any way modify their relationship to any resource type. The 13 Core attributes are:
CreatedTime
Creator
DeletedTime
Description
DetectedRulesList
DisplayName
ExpectedRulesList
ExpirationTime
Locale
MVObjectID
ObjectID
ObjectType
ResourceTime
You should not delete your schema resources while you still have auditing requirements for these resources.
In FIM, it can be helpful to make some regular expressions case insensitive. You can ignore case within a group by using ?!:
. For example, for Employee Type, use
^(?!:contractor|full time employee)%
.
The Member attribute exposed to the synchronization engine is actually mapped to ComputedMembers. It is a combination of criteria-based members and manually selected members. Even if you add all three attributes, (Filter, ExplicitMembers, and ComputedMembers), the dynamic calculation of the member attribute does not occur for resource types other than for group and set.
In FIM, you can enter strings with leading and trailing spaces, but the FIM system ignores those spaces. If you submit a string with a leading and trailing space, the synchronization engine and Web services ignores those spaces.
Empty strings are not equal to null in this release of FIM. Empty string input is regarded as a valid value. Not present is regarded as a null.
The following workflows are shipped with FIM 2010 and should not be deleted:
Expiration Workflow
Filter Validation Workflow for Administrators
Filter Validation Workflow for Non-Administrators
Group Expiration Notification Workflow
Group Validation Workflow
Owner Approval Workflow
Password Reset Action Workflow
Password Reset AuthN Workflow
Requestor Validation With Owner Authorization
Requestor Validation Without Owner Authorization
System Workflow Required for Registration
You should not run two or more ApprovalActivities in parallel. Doing so may cause the request to get stuck in the authorizing phase. For multiple approvals, either include a larger list of approvers in the approval or sequence the two activities back-to-back.
Avoid using activities that modify the FIM resources, such as the Function Evaluator Activity, as part of the workflows in authorization workflows. Because the request has not been committed while in the authorization point of processing, any modifications performed to the identity information can be applied despite the request being possibly rejected.
The objective of FIM is to process requests that can be initiated by various FIM clients such as the FIM synchronization service and the self-service components according to your configured business policies. By design, each FIM service instance belongs to a logical group that consists of one or more FIM service instances, which is also known as FIM service partition. If you have only one FIM service instance deployed to handle the all requests, it is possible that you experience processing latencies. Some operations can even exceed the default timeout values that are appropriate for self-service operations. FIM service partitions can help you to address this issue. For additional information see Understanding FIM Service Partitions.