I would reiterate what Indra said as well.
I’ve implemented High-Availability for our own organization and it’s pretty straight forward from the overview. I would read over the the references again and review/research each component to understand the purpose of each.
High Level though and ignoring things outside of UiPath you’ll be looking at
- 2 or more Orchestrator Nodes
- Those Orchestator Nodes would be kept in sync the RESP for communication (i.e. Redis) which is only supported if you use UiPath High Availability Add-On, but others are compatable. For Redis this would be 3 or more nodes in a cluster
- SQL Server Cluster with Always On
- Load-balancer in front of Orchestrator Cluster and Redis Cluster
Beyond that its more about your architectural design if you are going for Uptime Availability, Recovery, quick Standby/Backup for Maintenance and other outages, etc.
I need to know where each component (except Robots) has to be installed.
I think most would see each ‘node’ of a component provisioned on its own VM/Server in a production environment.
How many workstations/server are we talking about?
As above, this is more about your design, requirements and other needs, budget, etc.
For example having a NLB in front of your Orchestrator Nodes is good… but what happens if the NLB fails? So then are you deploying a second NLB in an Active/Standby pair? Similar to Orchestrator… 2 or more nodes are needed for HA… but is this about Scaling or is it about up time? Are you looking for quick disaster recovery, geo-located data centres etc. Similar examples for your Storage, Redis, SQL Server, etc.
Most probably wouldn’t put a DC and Orchestrator on the same hosts… assuming you have a DC other things in your network/organization depend on it to be available… But technically speaking, there is nothing stopping you from doing so, it just not recommended.
Again this will be highly dependent on your architectural needs. By default you are looking at a Local File System and with multiple Orchestrator nodes, you’ll want some form of central repository for your Storage… a simple solution is a Network File Share of sorts.
That said Orchestator supports multiple Storage Providers such as AWS S3 Buckets, or MinIO, so you have options. Just keep in might however you tackle it the File Read/Write should be kept in sync so you don’t get a disconnect between the File Objects and the meta data which is stored in the SQL Database.
I would recommend looking over the documentation for each component and get a better feeling for each. As your inquiry is very broad and ambiguous for the level of detail and time that would needed, if you have some specific pointed questions, I’m sure the community could offer some further insight.
My last tidbit of advice would be to reach out the UiPath’s Support or Sales team, they would be able to put you in contact with someone that is on the technical side to help guide you through the technical needs and implementation details.