# Advice on Hardware Configuration **URL:** \ **Category:** Ignition\ **Tags:** perspective\ **Created:** [July 25, 2024, 6:31am UTC](https://forum.inductiveautomation.com/t/advice-on-hardware-configuration/93729 "2024-07-25T06:31:01Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![Corne\_de\_Jonge](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/corne_de_jonge/32/46823_2.png) [@Corne\_de\_Jonge](https://forum.inductiveautomation.com/u/Corne_de_Jonge)\ **Post date:** [July 25, 2024, 6:31am UTC](https://forum.inductiveautomation.com/t/advice-on-hardware-configuration/93729/1 "2024-07-25T06:31:01Z")
Hi everyone, We're currently evaluating different PC configurations for running SCADA Ignition Perspective and would appreciate some advice on the best hardware setup. We are considering three options, listed in order of increasing price: 1. **Intel i5-1135G7 (4 cores), 32GB RAM, Intel Iris Xe Graphics** 2. **Intel i5-1335UE (10 cores), 32GB RAM, Intel UHD Graphics** 3. **Intel i7-1365UE (10 cores), 32GB RAM, Intel Iris Xe Graphics** All systems come with SSD storage. The Ignition gateway will have eight fixed client PCs connected to it. We also anticipate adding mobile clients, potentially increasing the total number of clients to around 15. We want some headroom for future growth. Our Perspective application is relatively large and includes a zoom and pan module from BIJC, which we've noticed requires some processing power to run smoothly. We anticipate having a tag count of about 150,000 to 200,000. Given these requirements, which hardware configuration would be recommended for the best performance and scalability? We find it challenging to estimate the necessary specs and would appreciate any insights or experiences you could share. ---
**Author:** ![Hayden\_Watson](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/hayden_watson/32/48792_2.png) [@Hayden\_Watson](https://forum.inductiveautomation.com/u/Hayden_Watson)\ **Post date:** [July 25, 2024, 6:38am UTC](https://forum.inductiveautomation.com/t/advice-on-hardware-configuration/93729/2 "2024-07-25T06:38:03Z")
This is a good article for the sizing of your servers. It will also depend a lot on how many tags you use and especially how quickly those tags are updated. > **[Ignition Server Sizing and Architecture Guide](https://inductiveautomation.com/resources/article/ignition-server-sizing-and-architecture-guide)** > > This guide is intended to provide some tips to help you determine the correct architecture depending on your requirements. It is important to note that any architecture that you come up with needs to be fully tested and verified. Throughout that... It is very difficult to accurately estimate the hardware, as so many factors change it, though I would personally think somewhere in the middle would be sufficient. ---
**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\ **Post date:** [July 25, 2024, 12:28pm UTC](https://forum.inductiveautomation.com/t/advice-on-hardware-configuration/93729/3 "2024-07-25T12:28:06Z")
That sizing guide is somewhat old--Perspective hadn't existed long when it was published--and doesn't provide satisfactory advice for Perspective users, IMNSHO. It really is only good for Vision applications. See this for more robust advice: > [@Microsoft Virtual Machine Suggestion](https://forum.inductiveautomation.com/t/microsoft-virtual-machine-suggestion/86787/5): > > After going through that sizing guide, I'd add another vCPU for every 4-8 Perspective clients (of pretty much any kind), a GB of gateway RAM for every 4-8 light-duty Perspective clients, and 2-4GB of gateway RAM for every heavy-duty Perspective client. "Light duty" would be single page Perspective apps running in Workstation, or Mobile, or single-tab browser sessions, displaying dashboards with real-time tags, up to an hour (-ish) of real-time trends, and pre-computed statistics for longer time… _ **Also** _, note that the sizing guides do _ **not** _ include database operations, but a system dedicated to just an Ignition gateway. (In production, databases should always be on separate systems, and sized for the quantity of data to be stored.) Edit: I just went back to see that the sizing guide has been updated for Cloud Edition, but still recommends client connections only suitable for Vision. It really needs to distinguish Vision clients from Perspective clients. Perspective clients put **way** more load on a gateway than Vision clients. ---
**Author:** ![dkhayes117](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/dkhayes117/32/68344_2.png) [@dkhayes117](https://forum.inductiveautomation.com/u/dkhayes117)\ **Post date:** [July 25, 2024, 12:48pm UTC](https://forum.inductiveautomation.com/t/advice-on-hardware-configuration/93729/4 "2024-07-25T12:48:54Z")
Interesting, I'm running +90 lightweight, Perspective clients and ~30 lightweight, Vision Clients. Only using 2-2.5 GB of RAM. ![image](https://us1.discourse-cdn.com/inductiveautomation/original/3X/9/5/952658ee5ebaf8a638fb7c4fc5ec5fdc492c231c.png) ---
**Author:** ![pturmel](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/pturmel/32/3295_2.png) [@pturmel](https://forum.inductiveautomation.com/u/pturmel)\ **Post date:** [July 25, 2024, 12:51pm UTC](https://forum.inductiveautomation.com/t/advice-on-hardware-configuration/93729/5 "2024-07-25T12:51:15Z")
How much charting in your lightweight Perspective clients? (My idea of lightweight includes scattered sparklines and realtime charting of **recent** data.) ---
**Author:** ![dkhayes117](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/dkhayes117/32/68344_2.png) [@dkhayes117](https://forum.inductiveautomation.com/u/dkhayes117)\ **Post date:** [July 25, 2024, 12:53pm UTC](https://forum.inductiveautomation.com/t/advice-on-hardware-configuration/93729/6 "2024-07-25T12:53:58Z")
Yes, I see your point. We have around 12-16 with any charting/tag history, the rest are mostly barcode/DB oriented.