A smart method for spark using neural network for big data

Apache spark, famously known for big data handling ability, is a distributed open-source framework that utilizes the idea of memory to process data. As performance spark mostly being affected by predominant configuration parameters, it challenging achieve optimal result from spark. The current practice tuning parameters ineffective, as performed manually. Manual large space and complex interactions with among parameters. This paper proposes more effective, self-tuning approach subject neural network called Smart method using (SSNNB) avoid disadvantages manual has selected five different sizes test approach. proposed increased speed around 30% compared default parameter configuration.