Azure Files scalability and performance targets (2023)

  • Article
  • 11 minutes to read

Azure Files offers fully managed file shares in the cloud that are accessible via the SMB and NFS file system protocols. This article discusses the scalability and performance targets for Azure Files and Azure File Sync.

The targets listed here might be affected by other variables in your deployment. For example, the performance of I/O for a file might be impacted by your SMB client's behavior and by your available network bandwidth. You should test your usage pattern to determine whether the scalability and performance of Azure Files meet your requirements. You should also expect these limits will increase over time.

Applies to

File share typeSMBNFS
Standard file shares (GPv2), LRS/ZRSAzure Files scalability and performance targets (1)Azure Files scalability and performance targets (2)
Standard file shares (GPv2), GRS/GZRSAzure Files scalability and performance targets (3)Azure Files scalability and performance targets (4)
Premium file shares (FileStorage), LRS/ZRSAzure Files scalability and performance targets (5)Azure Files scalability and performance targets (6)

Azure Files scale targets

Azure file shares are deployed into storage accounts, which are top-level objects that represent a shared pool of storage. This pool of storage can be used to deploy multiple file shares. There are therefore three categories to consider: storage accounts, Azure file shares, and individual files.

Storage account scale targets

Storage account scale targets apply at the storage account level. There are two main types of storage accounts for Azure Files:

  • General purpose version 2 (GPv2) storage accounts: GPv2 storage accounts allow you to deploy Azure file shares on standard/hard disk-based (HDD-based) hardware. In addition to storing Azure file shares, GPv2 storage accounts can store other storage resources such as blob containers, queues, or tables. File shares can be deployed into the transaction optimized (default), hot, or cool tiers.

  • FileStorage storage accounts: FileStorage storage accounts allow you to deploy Azure file shares on premium/solid-state disk-based (SSD-based) hardware. FileStorage accounts can only be used to store Azure file shares; no other storage resources (blob containers, queues, tables, etc.) can be deployed in a FileStorage account.

AttributeGPv2 storage accounts (standard)FileStorage storage accounts (premium)
Number of storage accounts per region per subscription250250
Maximum storage account capacity5 PiB1100 TiB (provisioned)
Maximum number of file sharesUnlimitedUnlimited, total provisioned size of all shares must be less than max than the max storage account capacity
Maximum concurrent request rate20,000 IOPS1100,000 IOPS
Throughput (ingress + egress) for LRS/GRS
  • Australia East
  • Central US
  • East Asia
  • East US 2
  • Japan East
  • Korea Central
  • North Europe
  • South Central US
  • Southeast Asia
  • UK South
  • West Europe
  • West US
  • Ingress: 7,152 MiB/sec
  • Egress: 14,305 MiB/sec
10,340 MiB/sec
Throughput (ingress + egress) for ZRS
  • Australia East
  • Central US
  • East US
  • East US 2
  • Japan East
  • North Europe
  • South Central US
  • Southeast Asia
  • UK South
  • West Europe
  • West US 2
  • Ingress: 7,152 MiB/sec
  • Egress: 14,305 MiB/sec
10,340 MiB/sec
Throughput (ingress + egress) for redundancy/region combinations not listed in the previous row
  • Ingress: 2,980 MiB/sec
  • Egress: 5,960 MiB/sec
10,340 MiB/sec
Maximum number of virtual network rules200200
Maximum number of IP address rules200200
Management read operations800 per 5 minutes800 per 5 minutes
Management write operations10 per second/1200 per hour10 per second/1200 per hour
Management list operations100 per 5 minutes100 per 5 minutes

1 General-purpose version 2 storage accounts support higher capacity limits and higher limits for ingress by request. To request an increase in account limits, contact Azure Support.

Azure file share scale targets apply at the file share level.

AttributeStandard file shares1Premium file shares
Minimum size of a file shareNo minimum100 GiB (provisioned)
Provisioned size increase/decrease unitN/A1 GiB
Maximum size of a file share
  • 100 TiB, with large file share feature enabled2
  • 5 TiB, default
100 TiB
Maximum number of files in a file shareNo limitNo limit
Maximum request rate (Max IOPS)
  • 20,000, with large file share feature enabled2
  • 1,000 or 100 requests per 100 ms, default
  • Baseline IOPS: 3000 + 1 IOPS per GiB, up to 100,000
  • IOPS bursting: Max (10000, 3x IOPS per GiB), up to 100,000
Throughput (ingress + egress) for a single file share (MiB/sec)
  • Up to 300 MiB/sec, with large file share feature enabled2
  • Up to 60 MiB/sec, default
100 + CEILING(0.04 * ProvisionedStorageGiB) + CEILING(0.06 * ProvisionedStorageGiB)
Maximum number of share snapshots200 snapshots200 snapshots
Maximum object name length3 (full pathname including all directories, file names, and backslash characters)2,048 characters2,048 characters
Maximum length of individual pathname component3 (in the path \A\B\C\D, each letter represents a directory or file that is an individual component)255 characters255 characters
Hard link limit (NFS only)N/A178
Maximum number of SMB Multichannel channelsN/A4
Maximum number of stored access policies per file share55

1 The limits for standard file shares apply to all three of the tiers available for standard file shares: transaction optimized, hot, and cool.

2 Default on standard file shares is 5 TiB, see Create an Azure file share for the details on how to create file shares with 100 TiB size and increase existing standard file shares up to 100 TiB. To take advantage of the larger scale targets, you must change your quota so that it is larger than 5 TiB.

3 Azure Files enforces certain naming rules for directory and file names.

File scale targets

File scale targets apply to individual files stored in Azure file shares.

AttributeFiles in standard file sharesFiles in premium file shares
Maximum file size4 TiB4 TiB
Maximum concurrent request rate1,000 IOPSUp to 8,0001
Maximum ingress for a file60 MiB/sec200 MiB/sec (Up to 1 GiB/s with SMB Multichannel)2
Maximum egress for a file60 MiB/sec300 MiB/sec (Up to 1 GiB/s with SMB Multichannel)2
Maximum concurrent handles per file, directory, and share root32,000 handles2,000 handles

1 Applies to read and write I/Os (typically smaller I/O sizes less than or equal to 64 KiB). Metadata operations, other than reads and writes, may be lower.

2 Subject to machine network limits, available bandwidth, I/O sizes, queue depth, and other factors. For details see SMB Multichannel performance.

3 Azure Files supports 2,000 open handles per share, and in practice can go higher. However, if an application keeps an open handle on the root of the share, the share root limit will be reached before the per-file or per-directory limit is reached.

Azure File Sync scale targets

The following table indicates which target are soft, representing the Microsoft tested boundary, and hard, indicating an enforced maximum:

ResourceTargetHard limit
Storage Sync Services per region100 Storage Sync ServicesYes
Sync groups per Storage Sync Service200 sync groupsYes
Registered servers per Storage Sync Service99 serversYes
Cloud endpoints per sync group1 cloud endpointYes
Server endpoints per sync group100 server endpointsYes
Server endpoints per server30 server endpointsYes
File system objects (directories and files) per sync group100 million objectsNo
Maximum number of file system objects (directories and files) in a directory (not recursive)5 million objectsYes
Maximum object (directories and files) security descriptor size64 KiBYes
File size100 GiBNo
Minimum file size for a file to be tieredBased on file system cluster size (double file system cluster size). For example, if the file system cluster size is 4 KiB, the minimum file size will be 8 KiB.Yes


An Azure File Sync endpoint can scale up to the size of an Azure file share. If the Azure file share size limit is reached, sync will not be able to operate.

Azure File Sync performance metrics

Since the Azure File Sync agent runs on a Windows Server machine that connects to the Azure file shares, the effective sync performance depends upon a number of factors in your infrastructure: Windows Server and the underlying disk configuration, network bandwidth between the server and the Azure storage, file size, total dataset size, and the activity on the dataset. Since Azure File Sync works on the file level, the performance characteristics of an Azure File Sync-based solution should be measured by the number of objects (files and directories) processed per second.

For Azure File Sync, performance is critical in two stages:

  1. Initial one-time provisioning: To optimize performance on initial provisioning, refer to Onboarding with Azure File Sync for the optimal deployment details.
  2. Ongoing sync: After the data is initially seeded in the Azure file shares, Azure File Sync keeps multiple endpoints in sync.


When many server endpoints in the same sync group are syncing at the same time, they are contending for cloud service resources. As a result, upload performance will be impacted. In extreme cases, some sync sessions will fail to access the resources, and will fail. However, those sync sessions will resume shortly and eventually succeed once the congestion is reduced.

To help you plan your deployment for each of the stages, below are the results observed during the internal testing on a system with a config

System configurationDetails
CPU64 Virtual Cores with 64 MiB L3 cache
Memory128 GiB
DiskSAS disks with RAID 10 with battery backed cache
Network1 Gbps Network
WorkloadGeneral Purpose File Server
Initial one-time provisioningDetails
Number of objects25 million objects
Dataset Size~4.7 TiB
Average File Size~200 KiB (Largest File: 100 GiB)
Initial cloud change enumeration80 objects per second
Upload Throughput20 objects per second per sync group
Namespace Download Throughput400 objects per second

Initial one-time provisioning

Initial cloud change enumeration: When a new sync group is created, initial cloud change enumeration is the first step that will execute. In this process, the system will enumerate all the items in the Azure File Share. During this process, there will be no sync activity i.e. no items will be downloaded from cloud endpoint to server endpoint and no items will be uploaded from server endpoint to cloud endpoint. Sync activity will resume once initial cloud change enumeration completes.The rate of performance is 80 objects per second. Customers can estimate the time it will take to complete initial cloud change enumeration by determining the number of items in the cloud share and using the following formulae to get the time in days.

Time (in days) for initial cloud enumeration = (Number of objects in cloud endpoint)/(80 * 60 * 60 * 24)

Initial sync of data from Windows Server to Azure File share:Many Azure File Sync deployments start with an empty Azure file share because all the data is on the Windows Server. In these cases, the initial cloud change enumeration is fast and the majority of time will be spent syncing changes from the Windows Server into the Azure file share(s).

While sync uploads data to the Azure file share, there is no downtime on the local file server, and administrators can setup network limits to restrict the amount of bandwidth used for background data upload.

Initial sync is typically limited by the initial upload rate of 20 files per second per sync group. Customers can estimate the time to upload all their data to Azure using the following formulae to get time in days:

Time (in days) for uploading files to a sync group = (Number of objects in server endpoint)/(20 * 60 * 60 * 24)

Splitting your data into multiple server endpoints and sync groups can speed up this initial data upload, because the upload can be done in parallel for multiple sync groups at a rate of 20 items per second each. So, two sync groups would be running at a combined rate of 40 items per second. The total time to complete would be the time estimate for the sync group with the most files to sync.

Namespace download throughput When a new server endpoint is added to an existing sync group, the Azure File Sync agent does not download any of the file content from the cloud endpoint. It first syncs the full namespace and then triggers background recall to download the files, either in their entirety or, if cloud tiering is enabled, to the cloud tiering policy set on the server endpoint.

Ongoing syncDetails
Number of objects synced125,000 objects (~1% churn)
Dataset Size50 GiB
Average File Size~500 KiB
Upload Throughput20 objects per second per sync group
Full Download Throughput*60 objects per second

*If cloud tiering is enabled, you are likely to observe better performance as only some of the file data is downloaded. Azure File Sync only downloads the data of cached files when they are changed on any of the endpoints. For any tiered or newly created files, the agent does not download the file data, and instead only syncs the namespace to all the server endpoints. The agent also supports partial downloads of tiered files as they are accessed by the user.


The numbers above are not an indication of the performance that you will experience. The actual performance will depend on multiple factors as outlined in the beginning of this section.

As a general guide for your deployment, you should keep a few things in mind:

  • The object throughput approximately scales in proportion to the number of sync groups on the server. Splitting data into multiple sync groups on a server yields better throughput, which is also limited by the server and network.
  • The object throughput is inversely proportional to the MiB per second throughput. For smaller files, you will experience higher throughput in terms of the number of objects processed per second, but lower MiB per second throughput. Conversely, for larger files, you will get fewer objects processed per second, but higher MiB per second throughput. The MiB per second throughput is limited by the Azure Files scale targets.

See also

  • Understand Azure Files performance
  • Planning for an Azure Files deployment
  • Planning for an Azure File Sync deployment
Top Articles
Latest Posts
Article information

Author: Carmelo Roob

Last Updated: 04/09/2023

Views: 5951

Rating: 4.4 / 5 (65 voted)

Reviews: 88% of readers found this page helpful

Author information

Name: Carmelo Roob

Birthday: 1995-01-09

Address: Apt. 915 481 Sipes Cliff, New Gonzalobury, CO 80176

Phone: +6773780339780

Job: Sales Executive

Hobby: Gaming, Jogging, Rugby, Video gaming, Handball, Ice skating, Web surfing

Introduction: My name is Carmelo Roob, I am a modern, handsome, delightful, comfortable, attractive, vast, good person who loves writing and wants to share my knowledge and understanding with you.