There are several differences between cloud and parallel databases, including load sharing, joins, query optimization, route planning, and optimization resources. The first difference is load sharing. At the same time, the database load sharing is balanced since the optimization algorithm works at the kernel level and the routing protocol determines the best route to find it. As a result, the load continues to be distributed on the server side. Moreover, in parallel database, table lookup is not necessary because the heuristic approach informs the server if the load balancing bit is set to one. The goal of load sharing strategies is to minimize the average transaction response time. In a parallel database, dynamic load sharing strategy is used for load sharing purposes. Xiaohu, Renqiang, Chujie, Yu (2013, p. 369) state that if at any time the cumulative performance of one component fails, then the cumulative performance of the rest of the components in the system must be redistributed to maintain the initial and final cumulative performance . performance of the entire system identical. Thus, it reduces the time required to resolve the query. Clients can make multiple requests at the same time. Unlike parallel database, cloud database collects all the resources into a pool. So whenever a resource is requested, the pool responds to that request and loads. it depends on the number of requests. Pitoura, Ntarmos, and Traintafillou (2012, p. 1315) argue that “range queries progress from the peer responsible for the lowest value in the range to the peer responsible for its highest value by following successor pointers” . On the contrary, in a parallel database, successive pointers are not used while all nodes are connected in a tree structure. Cloud database mainly deals with software as a service, in ...... middle of paper ......e insecure mode, many applications do not use this protocol. UDP collectively aggregates all packets through the channel and completely blocks resources. Additionally, UDP does not allow reliable data transfers because reliability costs more. For example, Li, Huang, Li, and Li (2013, p. 363) explain that after combining real-time data in the industrial field with cloud storage system organization and management technologies, the constraint in Real-time memory and storage index mechanism can be realized. . Jurisdiction Management deals with branch prediction technology; therefore, it automatically retrieves data before it is needed, speeding up the process. On the other hand, capacity distribution mainly concerns the processor cache. So when it encounters the same set of queries, it automatically retrieves the word. As a result, UDP ensures data transfer