Virtualization
Hi
We have 8 client and three redundant server (ASPECT, Connectivity, Domain), we want upgrade software and hardware and reduce hardware system by Virtualization into one system for clients and one system for servers. how should be select server specification such as hard storage, cpu, RAM, graphic and other specification?
thanks
Answers
Virtual servers (hosts) used for hosting thin client sessions should run on as fast (in terms of GHz not number of cores) processors as possible while hosts used by servers need less GHz.
Still, thin client operator workplaces cannot compete with rich clients in terms of operator satisfaction. Too much power is lost by removing the graphics hardware acceleration and furnishing the Microsoft Windows RDP compression/decompression. Multiple screens and dense process graphics aggravates the problem. However, some sites use thin client technology with success.
Count two virtual processor cores for regular servers while RDP servers should have at least one per workplace plus some extra. Some application servers have higher requirements.
There is a dedicated users guide for virtualization, read it.
To maintain three possible executing server locations, three hosts are required. Most virtual projects I get to see run 1oo2 AS.
Still, thin client operator workplaces cannot compete with rich clients in terms of operator satisfaction. Too much power is lost by removing the graphics hardware acceleration and furnishing the Microsoft Windows RDP compression/decompression. Multiple screens and dense process graphics aggravates the problem. However, some sites use thin client technology with success.
Count two virtual processor cores for regular servers while RDP servers should have at least one per workplace plus some extra. Some application servers have higher requirements.
There is a dedicated users guide for virtualization, read it.
To maintain three possible executing server locations, three hosts are required. Most virtual projects I get to see run 1oo2 AS.
Virtualization is an excellent idea for reducinng the number of physical servers required. However, it does require a new skill set if you have not used VMware before.
Upgrading your 800xA system is not a task for beginners. Planning the correct Hardware to run new virtual servers, ordering the correct upgrade components, installing the software and converting the graphics are complex tasks that require some experience. And then you have to get the system installed onto the plant without disrupting production. Upgrading an 800xA system is not "plug and go"
You will find the upgrade much easier, and you will save weeks of effort, and disruption, if you get some expert assistance to guide you through the process and to help complete the upgrade.
You are welcome to contact me off forum to discuss, or talk to your local ABB Group.
Upgrading your 800xA system is not a task for beginners. Planning the correct Hardware to run new virtual servers, ordering the correct upgrade components, installing the software and converting the graphics are complex tasks that require some experience. And then you have to get the system installed onto the plant without disrupting production. Upgrading an 800xA system is not "plug and go"
You will find the upgrade much easier, and you will save weeks of effort, and disruption, if you get some expert assistance to guide you through the process and to help complete the upgrade.
You are welcome to contact me off forum to discuss, or talk to your local ABB Group.
> "Virtualization into one system for clients and one system for servers"
This is a BAD idea.
You currently have 1-o-o-3 redundancy with physical servers and software. You're swapping it for non redundant hardware.
You need 3 physical servers to run the 3 Aspect Servers on separate hardware. Then you distribute the domain controllers and connectivity servers through this hardware so that no redundant service runs on the same hardware. Then you distribute the hardware around the plant so when the server room catches fire your system keeps running.
This is a BAD idea.
You currently have 1-o-o-3 redundancy with physical servers and software. You're swapping it for non redundant hardware.
You need 3 physical servers to run the 3 Aspect Servers on separate hardware. Then you distribute the domain controllers and connectivity servers through this hardware so that no redundant service runs on the same hardware. Then you distribute the hardware around the plant so when the server room catches fire your system keeps running.
Hi!
Refer to these documents for guidance:
3BSE056141-610_A_en_System_800xA_6.1_Virtualization_with_VMware_vSphere_ESXi
3BSE082294-610_A_en_System_800xA_6.1_Virtualization_with_Microsoft_Hyper-V
BR
Add new comment