Alfresco architecture for 1TB of data

[miroslav]

Hi,

I would like to import 1TB of data (mostly pdf) to Alfresco Community 7. There will be around 10 active users working wiith platform at the same time.

Does the Starter deployment architecture of alfresco whiteparers apply to these cases? (one server for repository/share/solr, one server for metadata)
Is it enough to separate the metadata server (PostgreSQL) with new version of search services?
What architecture and sizing would you choose?

Thank you!

[angelborroy]

In order to provide a better approach, the number of documents (rather than storage volume) would help.

How many documents are expected to live in the Repository?

What is the ingestion rate you are expecting in the future?

[miroslav]

@angelborroy Hi, thank you for answer. Thre will be approximately 2M pdf documents with an increase of about 20k per year. Some data can be archived (or stored on slower HDD). 

[angelborroy · Accepted solution]

I guess the suggested base architecture should be enough for that:

* 1 server for Repo (including Transform Service, as it's not relevant for your use case)

* 1 server for Search Services (SOLR)

* 1 server for Database