Scalability is the ability of a system to handle growing amount of load without degrading performance. Consider you have system which runs with 100 users efficiently. Let’s say over a period of time your load increases to 500 users, your system still has the ability to handle the load and provide the same efficiency.
You can achieve scalability by two ways either you can Scale-up or Scale-out. In scale-up we have only one machine and we increase the processing power of the machine by adding more processor, more ram, more hard disk etc. So if your load increases you add more ram , more processor etc , but you do not add extra physical machines.
In Scale-out, as your load increases you add more computers and you have load balancers in front of those machines to distribute load appropriately.
Taken from the book important Dotnet interview questions by ShivprasadKoirala. You can also watch this wonderful video of