Review and Update MongoDB for Windows Install Guide (DOCS-11434)

[DOCS-6329] Recommended NTFS cluster size Created: 06/Oct/15  Updated: 16/Jun/18  Resolved: 15/Jun/18

Status: Closed
Project: Documentation
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Sub-task Priority: Major - P3
Reporter: Andrew Ryder (Inactive) Assignee: Anthony Sansone (Inactive)
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
Participants:
Days since reply: 5 years, 34 weeks, 5 days ago

 Description   

It would be nice to have recommendations on Windows NTFS cluster sizes, given we have some things to say about the configuration of Linux file systems. Windows tends to be more plug & play but cluster size will certainly have a small impact.

It is likely we just recommend 4K as a blanket statement, which would suit MMAPv1, but there might be more range in WiredTiger. Not sure what would be best there.



 Comments   
Comment by Githook User [ 15/Jun/18 ]

Author:

{'username': 'atsansone', 'name': 'Anthony Sansone', 'email': 'tony.sansone@mongodb.com'}

Message: (DOCS-6329): Updated default NTFS allocation unit size.
Branch: v3.6
https://github.com/mongodb/docs/commit/176307dd4129bf907b61b2ad2aa633ca66b34f35

Comment by Githook User [ 15/Jun/18 ]

Author:

{'username': 'atsansone', 'name': 'Anthony Sansone', 'email': 'tony.sansone@mongodb.com'}

Message: (DOCS-6329): Updated default NTFS allocation unit size.
Branch: master
https://github.com/mongodb/docs/commit/b472c0a76504bfbdd0b2de8fbb4c8b493b0b375e

Comment by Tom Li [ 13/Jun/17 ]

I am just wondering if we are still planning to improve the doc Operations Checklist for Windows to have recommendations on Windows NTFS cluster size. I have a customer who is asking the same thing and it would be nice to be able to refer the customer to our official docs for this recommendation.

Comment by Michael Cahill (Inactive) [ 08/Oct/15 ]

I think 4KB makes sense for WiredTiger as well – we will write data in multiples of 4KB, so a smaller size doesn't make sense, and I assume there is overhead if we write 4KB blocks and the cluster size is larger.

Generated at Thu Feb 08 07:52:05 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.