<Disclaimer>This is personal notes of what I retained during the session. This can be incomplete, partially right or wrong. It is just part of the notes I took and what retained my attention. Nothing prevents the user to get more information on their favorite web site.</Disclaimer>

New guidance issued by Microsoft in June with a possibility to scale up to 4TB (or beyond) for the content database. It requires 2 IOPs per GB, split in several content database to reduce the I/O contention. It is also recommended to upgrade to SP1.

The unlimited size support is for record center and document center only. On the other side, avoid using alerts, workflows and item-level security.

To achieve the needed performance, it is better to have more and faster disks, rather than larger and slower disks. Isolate the TEMPDB on its own RAID volume, the same apply for the log files. RAID  5 is acceptable, but RAID 1+0 is recommended. For a large number of items, row density is important in the content database. For the search crawl, the recommended limit is at 25 million.

To estimate the storage requirements, you need to know the number and types of documents, because it impacts the crawler performances. The average file size is also important. Documents large than 16 MB are not crawled by SharePoint, by default. Can be increased to a max of 64MB. The number of versions per document has to be taken in account.

<snip>A lot of formulas are presented, check the video and the powerpoint deck to get the details</snip>

To test the storage performance, there are two free tools, SQLIO (from Microsoft) and IOMeter. The testing files should be larger than the SAN cache and testing should be done for an hour or more. Storage should be dedicated.

Recovery Point Objective (RPO) is how much I can loose and the Recovery Time Objective (RTO) is how much time I can wait for the recovery. We need to be monitor in order to be sure we can achieve the RPO and RTO. If RPO states 2 hours but backup take 4 hours => FAIL.