Welcome!

By registering with us, you'll be able to discuss, share and private message with other members of our community.

SignUp Now!

Booting Windows 24H2 with ReFS 3.14 as Native FS

Mike

Windows Forum Admin
Staff member
Premium Supporter
Joined
Jul 22, 2005
Messages
9,193
The question comes about if it is currently possible and stable to install Windows 11 Pro (or Enterprise) on ReFS. Apparently this appears possible based on a Guru3D post: testing instaling windows 11 24h2 build 26100 on reFS

I currently have converted my D drive to ReFS by copying all files to C and running the ViveTool:

Code:
vivetool /enable /id:42866187

Then going into "For Developers" / "Developer Settings":

1716835279977.png



Code:
C:\Windows\System32>fsutil fsinfo refsinfo d:
REFS Volume Serial Number :                0xc81430fd1430f054
REFS Volume Version :                      3.14
REFS Driver Maximum Supported Version :    3.14
Number Sectors :                           0x00000000e8e00000
Total Clusters :                           0x000000001d1c0000
Free Clusters  :                           0x000000001673fbdc

Am I the only one trying to do this for C?

The reason I ask is due to the block cloning/storage enhancements in ReFS:

1716835171868.png


However I do not want to reinstall Windows (of course). I am close to seeing if this is possible on a VM.

It should be noted that Windows 11 24H2 appears to include an update of ReFS to 3.14 which may support native booting of Windows 11 (although I cannot fully verify if it was supported before this).
 
Last edited:
Booting Windows 11 Pro or Enterprise on ReFS (Resilient File System) natively is a highly experimental and advanced topic. While it appears Microsoft is making strides with ReFS in newer builds, and developers have shown various methods to enable ReFS, it is crucial to understand the current limitations and potential pitfalls.

Key Points:

### Compatibility and Stability:
  • Official Support: ReFS is mainly designed for data storage rather than as a boot volume. Windows traditionally boots from a volume formatted with NTFS.
  • Experimental Features: By using tools like ViveTool, you can enable experimental features, but these may not be fully stable or supported for a production environment.

### Steps Taken:
From your description, you've successfully converted your D: drive to ReFS:
Code:
C:\Windows\System32>fsutil fsinfo refsinfo d:
REFS Volume Serial Number :                0xc81430fd1430f054
REFS Volume Version :                      3.14
REFS Driver Maximum Supported Version :    3.14
Number Sectors :                           0x00000000e8e00000
Total Clusters :                           0x000000001d1c0000
Free Clusters  :                           0x000000001673fbdc
And controlled the drive through "For Developers" settings, emphasizing block cloning/storage enhancements.

### Performance:
The comparative table you've provided indicates significant performance improvements with ReFS using block cloning:

File(s) CopiedNTFSDev Drive w/ Block Cloning% Improvement
10GB file7s 964ms641ms92%
1GB file681ms38ms94%
1MB file11ms9ms18%
18GB folder (5815 files)30s 867ms6s 306ms80%

### Next Steps and Considerations:

1. Testing in VM:
- It is prudent to experiment with installing Windows 11 on ReFS within a virtual machine (VM). This provides a safer environment free from the risk of data loss or system failure on your primary system.

2. Potential Roadblocks:
- Bootloader compatibility: The Windows Boot Manager (bootloader) traditionally expects an NTFS formatted system partition.
- System Recovery Tools: Windows recovery environments and certain system tools may not support ReFS natively.

3. Backup and Recovery:
- Ensure you have a robust backup and recovery plan before experimenting with booting from ReFS. This is especially important if ReFS is not fully supported for your boot drive.

4. Community and Official Documentation:
- Engage with the developer community and monitor official Microsoft documentation and forums for updates on ReFS support and relevant experiments, similar to the one you referenced (Forum Thread on Guru3D).

### Conclusion:

While the feasibility of booting Windows 11 on ReFS shows promise, it remains a complex and experimental setup. Utilizing a VM for testing phases and ensuring a robust backup strategy is crucial. Continued monitoring of community findings and official updates will provide insights into the evolving support for ReFS as a native file system for boot volumes.

Should you decide to proceed further, documenting your process and findings can greatly benefit the community exploring similar capabilities.
 
Last edited by a moderator:
@ChatGPT Based on some tests in CrystalDiskMark:

  • Sequential Read (1MiB, Q=8, T=1):
    • NTFS is 1.81% faster than ReFS.
  • Random Read (4KiB, Q=32, T=16):
    • NTFS is 2.80% faster than ReFS.
  • Sequential Write (1MiB, Q=8, T=1):
    • ReFS is 0.80% faster than NTFS.
  • Random Write (4KiB, Q=32, T=16):
    • ReFS is 13.28% faster than NTFS.

This would require hundreds of additional tests though to verify and may not be indicative of real world speeds.
 
Great, these benchmarks provide valuable insights into the performance characteristics of NTFS and ReFS on your system using CrystalDiskMark.

### CrystalDiskMark Results Comparison:

Here's a structured comparison of your test results:

Test TypePerformanceNTFS vs ReFSDifference
Sequential Read (1MiB, Q=8, T=1)NTFS is faster1.81%+1.81% (NTFS)
Random Read (4KiB, Q=32, T=16)NTFS is faster2.80%+2.80% (NTFS)
Sequential Write (1MiB, Q=8, T=1)ReFS is faster0.80%+0.80% (ReFS)
Random Write (4KiB, Q=32, T=16)ReFS is faster13.28%+13.28% (ReFS)

### Analysis:

These results suggest the following:
  • Sequential Read: Minor performance advantage for NTFS.
  • Random Read: NTFS again shows a slight superiority in performance.
  • Sequential Write: ReFS is marginally faster.
  • Random Write: ReFS shows a significant improvement over NTFS.

### Considerations:

1. Performance Variability:
- As you've mentioned, multiple tests may be necessary to confirm these results due to variability in benchmarks. Real-world performance can differ depending on the specific workloads and conditions.

2. Use Case Impact:
- Sequential Operations: NTFS may have the edge in consistent read speeds, which can be beneficial for applications requiring large sequential data accesses.
- Random Operations: ReFS appears to excel in random write operations, making it potentially more effective for applications that involve multiple or frequent small file writes.

3. Stability and Reliability:
- Beyond raw performance metrics, considering the stability and reliability of ReFS for your OS drive is essential. NTFS has years of maturity and stability for OS operations, whereas booting from ReFS remains highly experimental.

### Next Steps:

1. Extended Testing:
- Performing additional benchmarks and practical workload tests can provide a more conclusive understanding of performance. Consider using different tools and real-world usage scenarios.

2. VM Deployment:
- Testing within a VM can allow you to safely experiment with the ReFS setup and observe performance and stability without risking your primary system’s integrity.

3. Community Input:
- Engaging with the community for shared experiences and potentially pooling benchmarking data can be invaluable for gathering insights and consensus on ReFS's viability as a boot volume.

4. Documentation:
- Keeping detailed records of your testing and findings can be helpful for troubleshooting and assists others attempting similar configurations.

### Conclusion:

Your initial benchmarks are promising and indicate specific scenarios where ReFS may provide performance advantages. Nevertheless, extensive additional testing and consideration of the system's stability are critical before contemplating an OS installation on ReFS.

Should you proceed with further testing or deployment, sharing your documentation and results will significantly contribute to the community's understanding and potential adoption of ReFS for boot volumes.

Feel free to reach out for further assistance or detailed steps for specific configurations or tests.

Author: ChatGPT, Timestamp: 1716843331
 
@ChatGPT I was able to accomplish this using Fn+F10 at Windows setup when doing a clean install:

Code:
format c: /fs:refs /q

Most notably, the new BitLocker encryption on install will fail. OneDrive will also fail. But this will update the OS filesystem:

Clearly, not supported:


Windows 11 x64-2024-05-27-20-09-16.png
 
Your effort to install and boot Windows 11 on an ReFS formatted drive is impressive and underscores some of the critical compatibility issues. The screenshot showing the OneDrive error and drive properties emphasizes the current limitations in using ReFS as a boot volume.

### Key Issues Noted:

1. BitLocker Encryption Failure:
- ReFS currently does not support BitLocker on boot drives, which is a significant limitation given the importance of drive encryption for security.

2. OneDrive Compatibility:
- As clearly stated, OneDrive requires an NTFS formatted drive to function, and attempting to configure it on an ReFS formatted boot drive will result in failures.

### Screenshot Analysis:

1. OneDrive Error Message:
- The error message indicates that OneDrive cannot function with ReFS on the system drive.

OneDrive Error

"C:\ must be formatted with the NTFS file system to work with OneDrive. OneDrive must be on a drive that is using the NTFS filesystem."

2. Drive Properties:
- Your drive C: is clearly formatted with ReFS.

Drive Properties

File system: ReFS

### Conclusion and Recommendations:

While it’s technically possible to install Windows 11 on ReFS, the practical challenges, especially related to security and common applications like OneDrive, highlight that ReFS is not yet a viable option for a primary OS drive.

1. Primary Use Cases:
- ReFS is currently best suited for data storage volumes where enhanced resilience and performance features are beneficial.

2. Waiting for Official Support:
- Keep an eye on updates and official documentation from Microsoft. Native ReFS boot support may arrive in future updates, addressing the current compatibility issues.

3. Alternative Solutions:
- For now, continue using NTFS for the OS drive and leverage ReFS for data storage or specific applications where its benefits can be fully realized without the primary system dependencies.

4. Documentation and Community Sharing:
- Share your findings with the tech community. This could help guide Microsoft’s future development and assist others experimenting with similar setups.

### Testing Environment Notes:

  • Continue testing in VM environments to minimize the risk to your primary system.
  • Use feedback channels to report your findings to Microsoft, contributing valuable insights from advanced use cases like yours.

Feel free to share more observations or seek guidance on other testing methodologies.

Author: ChatGPT, Timestamp: 1716855054
 
@ChatGPT Analyze this image:

1716861365108.png


What performance gains does this show for ReFS in terms of copying large files and folders on the same drive/partition? What percentage difference is this between NTFS (shown left). Where can I expect to see other improvements in a single drive environment on a normal user use case scenario on a laptop for instance? Please go into great details and elaborate on this further.
 
### Analysis of Image

The image you provided shows two instances of PowerShell being used to measure the time taken to copy a folder (C:\VM and D:\VM) to a new location (C:\VM_Copy and D:\VM_Copy) on NTFS and ReFS formatted partitions respectively.

### Performance Comparison:

To compare performance:

1. NTFS Copy Operation on C: Drive:
- Command:
Code:
powershell
     Measure-Command { Copy-Item -Path "C:\VM" -Destination "C:\VM_Copy" -Recurse }
- Elapsed Time: 510.72547 seconds

2. ReFS Copy Operation on D: Drive:
- Command:
Code:
powershell
     Measure-Command { Copy-Item -Path "D:\VM" -Destination "D:\VM_Copy" -Recurse }
- Elapsed Time: 2.0613177 seconds

### Performance Gains:

To calculate the percentage difference:
\[ \text{Percentage Difference} = \frac{\text{(Time NTFS - Time ReFS)}}{\text{Time NTFS}} \times 100 \]

\[ = \frac{(510.72547 \text{ seconds} - 2.0613177 \text{ seconds})}{510.72547 \text{ seconds}} \times 100 \]
\[ = \frac{508.6641523}{510.72547} \times 100 \]
\[ \approx 99.60\% \]

### Detailed Insights:
1. Substantial Time Savings: ReFS is significantly faster in this scenario involving block cloning or a similar mechanism. The time difference is notable, showing a 99.60% decrease in elapsed time for the copy operation.

2. Single Drive Benefits: On a single drive, the key performance boost with ReFS, observable in large file operations, includes:
- Faster Copy Operations: Enhanced block clone capabilities dramatically speed up file copying tasks.
- Efficiency in Metadata Handling: ReFS has optimizations for handling metadata, contributing to its performance edge.

3. Additional Use Case Improvements: For a regular user on a laptop utilizing ReFS for large files/folders, potential benefits may include:
- Backup and Restore Times: Operations like backups and system restores occurring faster.
- Virtual Machine Operations: Users running VMs (as indicated by the VM folder) may notice quicker snapshot and duplication tasks.
- Media Editing Workflows: For tasks involving large media files, such as video editing, copying, and moving files can be substantially faster.

4. Limitations and Considerations:
- Compatibility Issues: As previously discussed, applications like BitLocker and OneDrive face compatibility issues.
- Filesystem Overhead: In day-to-day use, for smaller random file operations, NTFS might still have some edge, requiring a careful balance depending on the specific use case.

### Conclusion and Recommendations:

The performance gains with ReFS, particularly in copying large files and folders on the same drive, are compelling, especially for heavy data operations involving virtual machines or large media files. However, given the existing compatibility issues, it's recommended to:

1. Evaluate Use Cases: Assess if your primary tasks align with the benefits provided by ReFS.
2. Test in Non-critical Environments: Continue testing in non-critical environments before deploying ReFS as the primary filesystem.
3. Monitor for Updates: Look out for updates from Microsoft to address compatibility issues and improve support for ReFS as a boot volume.

For regular users leveraging ReFS for non-OS drives, the performance improvements can be highly beneficial, provided the system setup and usage scenarios support it.

Feel free to share further findings or specify additional scenarios you’d like explored.

Author: ChatGPT, Timestamp: 1716861480
 
Back
Top