Advantages And Disadvantages Of Static Load Balancing

2.5 Load Balancing:
Load balancing (Sudhan and Jagdish, 2015) Load balancing is the process of distributing the load among various nodes of a distributed system to improve both resource utilization and job response time while also avoiding a situation where some of the nodes are heavily loaded while other nodes are idle or doing very little work. Load balancing ensures that each processor in the system and each node in the network carry almost the same amount or equal labor at any moment of time. An efficient balancing algorithm should avoid overloading or under loading any particular node in the system. Building a load balancing algorithm should take care of issues such as security, reliability, productivity, etc. The primary goal of balancing

once the load is allocated to a node that can not be transferred to another node. This method requires fewer connections and thus reduces the execution time, which enhances performance and minimize the communication delays. The disadvantages of static load balancing are that it is very difficult to compute a-priori execution time, the process allocation cannot be changed during execution and Sometimes there are communication delays that vary in an uncontrollable way for some problems the number of steps to reach a solution is not known in advance (Hendra and Yudi, 2009).
There are three types of static load balancing algorithms: round robin, randomized algorithm, central manager and threshold algorithm. 2.6.1.1 Round Robin Algorithm:
In this algorithm load is distributed equally to all nodes in the network. Work load is distributed in round robin order, where equal load is assigned to each node in circular order without any priority and will be back to the first node if the last node has been reached. Each node maintains its load index independent of allocations from remote node.
Round robin is easy to implement, gives best performance for special purpose applications. It does not require inter process communication . It cannot give expected result in general case and when the jobs are of unequal processing time . This lead to reduce the throughput becuase of larger waiting time and response time (Shengsheng, et

It is only possible during the actual program execution when it becomes apparent how much work is being assigned to the individual processor. In order to retain efficiency, the imbalance must be detected and an appropriate dynamic load balancing strategy must be devised (Luque, et al., 1995).
Dynamic load balancing techniques depend on recent system load information and determine the job assignments to server nodes at run time. In case of dynamic approach, the load balancing decisions are based on the current state of the system and hence work loads are allowed to shift dynamically from an overloaded node to an under-loaded node to get faster response from the server nodes. This ability to respond to variations in the system is the main advantage of the dynamic load balancing. But, since the load balancer has to continuously monitor the current system load on all the nodes, it becomes an extra overhead as monitoring consumes CPU cycles. So, a proper decision must be taken when to invoke the monitoring operation by the load balancer (Beaulah, et al.,