Load Sharing Vs Load Balancing for Dummies

The term load balancing and load sharing are used interchangeably which is not correct. The overall functioning overlaps with each other but at one stage there is one significant difference.
The reason for usage of interchangeability is due the mechanism deployed at various context and term associated with them such HSRP, VRRP, GLBP, NAT or DNS based load sharing, equal/unequal cost path load sharing, ether channeling, reverse proxy, round robin, weighted queue etc.

Let take look at below explanation.

Load balancing and load sharing are both techniques for managing traffic and resources, but they work in different ways and have different benefits:

Load Sharing 

The traffic is distributed across multiple processing resources (such as multiple links/paths, servers, routers etc.) although no attempt is made to ensure that an even amount of parameters (such as load, bandwidth, delays, availability etc.) are utilized on each resources.

For e.g. connection A and connection B could each process 50% of the incoming/outgoing traffic although the bandwidth utilization on connection A could be far higher than the bandwidth utilization on connection B.

Typically traffic is directed to connection A or connection B by a round robin scheduler, although it can also be done by feeding either the source/destination MAC address, source/destination IP address, source/destination port number into an algorithm.

Load balancing

Load balancing goes one step further and looks at the bandwidth utilization before deciding which connection should be used to carry the traffic. The load balancing mechanism can usually take into account other factors such as delay, bandwidth utilization or link availability itself when deciding which connection to use.

Depending upon the load sharing or load balancing mechanism it is sometimes possible to do un-equal load sharing/balancing or a weighting when deciding which connection should be used.