===== System requirements =====

System requirements, software prerequisites, and other recommendations are described in a separate document [[https://host.pathomation.com/pdf/requirements-2019.pdf|"System Requirements"]] (but probably not as elaborate as in that document)

==== On-premise installation ====

^ Component      ^ Minimum       ^ Suggested          ^
| CPU cores    | 4     | 16        |
| RAM    | 8GB | 16GB |
| Disk    | 250GB HDD | 500GB SSD |
| Cache disk    | 100GB HDD     | 1TB HDD |
| Network interface    | 100MBps | 1Gigabit|

SQL Server can also be installed in a separate machine. The requirements for the computer hosting only SQL Server are the following:

^ Component      ^ Minimum       ^ Suggested          ^
| CPU cores    | 4     | 8        |
| RAM    | 8GB | 16GB |
| Disk    | 250GB HDD | 500GB SSD |
| Network interface    | 100MBps | 1Gigabit|

=== Additional comments regarding SQLServer ===
SQL Server Express can be used but poses two limitations which have to be taken into account:
  - The maximum database size cannot exceed 10GB.
  - Only one CPU core is used to execute queries.

==== Installation on an AWS Virtual Machine ====

When installing PMA.core on AWS, it is highly recommended to use RDS to host SQL Server instead of installing it on the VM that PMA.core will be installed.

^ Component ^ Minimum ^ Suggested ^
| PMA.core VM size| t3.large | t3.xlarge 	|
| SQL Server RDS VM size| t3.medium | t3.large |
| Cache disk | 100GB EBS | 1TB EBS |

Additionally, slides can be stored either on local disks attached to the VM or in S3 buckets.
==== Installation on an Azure Virtual Machine ====

When installing PMA.core on Azure, it is highly recommended to use the Azure SQL Server service to host SQL Server instead of installing it on the VM that PMA.core will be installed.

^ Component ^ Minimum ^ Suggested ^
| PMA.core VM size| D2as v5 | D4as v5 |
| Azure SQL Server RDS VM size| A2m v2 | A4 v2 |
| Cache disk | 100GB standard HDD | 1TB standard HDD |

Additionally, slides can be stored either on local disks attached to the VM or in Azure Blob Storage.
When possible, it is recommended to prefer [[https://docs.microsoft.com/en-us/azure/storage/blobs/data-lake-storage-introduction|Azure Data Lake Storage Gen2]] over plain Azure Blob Storage, as this type of service performs directory management operations, such as moving and renaming slides and directories, a lot faster.