SOLVED

5000 Threshold on Mondern List & Librariers is it still there?

Copper Contributor

Hi,

 

I try to find out what is currently the actual status or what is supported from Microsoft regarding the 5000 Threshold. I'm talking about modern Lists/Libraries and sorting filtering in the UI. 

 

Do we still have the limitations with 5000 Threshold?

 

I know there is the User Voice topic and I saw the video from Ignite 2018 regarding Large Lists, Indexing and paging. 

 

The User Voice is still on working on it and I found no official message from Microsoft.

 

@Vesa Juvonen @Mark Kashman 

4 Replies

@Benjamin Stierle 
Short answer: Yes
Long answer: this limitation is still present but use of indexed columns reduces the impact. The topic has a high vote on uservoice: https://sharepoint.uservoice.com/forums/329214-sites-and-collaboration/suggestions/8463457-remove-th...
AFAIK there is no item on the roadmap.
Paul

best response confirmed by Benjamin Stierle (Copper Contributor)
Solution

@Benjamin Stierle 

A lot of thought went into this article, "Living Large with Large Lists and Large Libraries" - of which is derived from the innovation of "predictive indexing. Plus our official support page on same, "Manage large lists and libraries".

 

Up to 30 million items can be kept in a SharePoint list. Predictive indexing helps as lists grow beyond 5000 items, SharePoint (where lists and libraries are stored) senses the fields used in views and sorts and automatically adds indexes without user intervention or throttles. The modern user experience is also optimized to use those indexes, when available – and to retrieve data in sets to avoid throttles and unavailability. Predictive indexing lets you use the full capacity of SharePoint lists and libraries while minimizing the need for active administration and performance throttles. When it detects a query that might result in thousands of records, SharePoint automatically adds the index on the fly – and we’re working on automatic indexing for queries of any size. In the modern user experience, library views are smarter about using queries and paging behind the scenes to grab records in manageable sets as the users moves through a large view, without needing to throttle the view completely.

@Mark Kashman 

 

thanks for the Answer.

 

So the improvents are already in production but we have to respect the current supported column types and the size regarding pre indexing

Hello Mark , I created a bulk (6000 items) in SP list via powershell. In list settings, it shows that I reached the limit but in my All Items view, there was no error and I can scroll all the way down to 6000 item without any problem. I am on Modern and I haven't done indexing yet. Is there a trigger for this once it reached 5000 limit? or would this be update in SharePoint online? Thank you.
1 best response

Accepted Solutions
best response confirmed by Benjamin Stierle (Copper Contributor)
Solution

@Benjamin Stierle 

A lot of thought went into this article, "Living Large with Large Lists and Large Libraries" - of which is derived from the innovation of "predictive indexing. Plus our official support page on same, "Manage large lists and libraries".

 

Up to 30 million items can be kept in a SharePoint list. Predictive indexing helps as lists grow beyond 5000 items, SharePoint (where lists and libraries are stored) senses the fields used in views and sorts and automatically adds indexes without user intervention or throttles. The modern user experience is also optimized to use those indexes, when available – and to retrieve data in sets to avoid throttles and unavailability. Predictive indexing lets you use the full capacity of SharePoint lists and libraries while minimizing the need for active administration and performance throttles. When it detects a query that might result in thousands of records, SharePoint automatically adds the index on the fly – and we’re working on automatic indexing for queries of any size. In the modern user experience, library views are smarter about using queries and paging behind the scenes to grab records in manageable sets as the users moves through a large view, without needing to throttle the view completely.

View solution in original post