Good day All,
Welcome back!!!
The other day someone asked me the difference between High Availability vs Fault tolerance in the cloud.
Administrators supporting on premises Application and Servers would just say oh High Availability is having multiple servers and if 1 goes down still you have another server.
If Fault tolerance then some one would say oh yeah Windows cluster one node is down and another node is available to Serve Application
Lot of Admins still get confused and always use both the terms as synonymous
Lets take a example and break this down.
Assume that some one from Application team comes and tells you that my Application is critical and my requirement is i need 2 Servers for better performance and SLA is that at-least 1 Server should be up and running
Solution: In this case the most likely solution is that you will build a 2 Node cluster and both terms High Availability and Fault Tolerance are same in this case
Lets take an another example:
Application team guy comes to you and tells that my Application is critical and my requirement is that at-least 2 Servers needs to be running all the time for better performance of the application and it should be highly available as well.
Solution: The likely solution in this case is building a 4 Node cluster with 2 Nodes as Active and 2 Nodes as Passive or 2 Node Active/ Active Cluster.
If you seen the requirement the SLA played a role in determining how you provide the solution and that differentiates the terms HA and FT.
If one of the VMware Admin reading this post probably will have a different perspective in terms of the terms HA and FT as this involves Virtual Machines
HA in terms of VMware would be If one ESX host goes down the virtual machine should be up and running in another ESX host
FT would be if Virtual machine does down another Virtual Machine should continue to Serve the Application running on the Virtual Machines.
The reason i wanted to post this now is because we are moving away from On premises to cloud and Application SLA's and cost will play a major role in building Servers and our understanding on both the terms should be very clear.
Hopefully this make sense and helps some one!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
If anyone else as any other thoughts free to reply on the post , Until next one you all have a good day!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Welcome back!!!
The other day someone asked me the difference between High Availability vs Fault tolerance in the cloud.
Administrators supporting on premises Application and Servers would just say oh High Availability is having multiple servers and if 1 goes down still you have another server.
If Fault tolerance then some one would say oh yeah Windows cluster one node is down and another node is available to Serve Application
Lot of Admins still get confused and always use both the terms as synonymous
Lets take a example and break this down.
Assume that some one from Application team comes and tells you that my Application is critical and my requirement is i need 2 Servers for better performance and SLA is that at-least 1 Server should be up and running
Solution: In this case the most likely solution is that you will build a 2 Node cluster and both terms High Availability and Fault Tolerance are same in this case
Lets take an another example:
Application team guy comes to you and tells that my Application is critical and my requirement is that at-least 2 Servers needs to be running all the time for better performance of the application and it should be highly available as well.
Solution: The likely solution in this case is building a 4 Node cluster with 2 Nodes as Active and 2 Nodes as Passive or 2 Node Active/ Active Cluster.
If you seen the requirement the SLA played a role in determining how you provide the solution and that differentiates the terms HA and FT.
If one of the VMware Admin reading this post probably will have a different perspective in terms of the terms HA and FT as this involves Virtual Machines
HA in terms of VMware would be If one ESX host goes down the virtual machine should be up and running in another ESX host
FT would be if Virtual machine does down another Virtual Machine should continue to Serve the Application running on the Virtual Machines.
The reason i wanted to post this now is because we are moving away from On premises to cloud and Application SLA's and cost will play a major role in building Servers and our understanding on both the terms should be very clear.
Hopefully this make sense and helps some one!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
If anyone else as any other thoughts free to reply on the post , Until next one you all have a good day!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Yes I agree in each scenarios
ReplyDeleteThis can be accomplished by creating a vm with ft enabled. This will solve the purpose and it's more cost effective