An assault on a network that overflows it with so many requests that regular traffic is either decelerated or entirely interrupted. Unlike a virus or worm, this can cause severe damage to databases. A Distributed Denial of service (DDos) attack can employ hundreds or even thousands of computers that have been previously flooded by HTTP GET packet. The massive amounts of data that collect over time which problematic to analyze using common database management tools. Big data includes activity logs (machine generated data) which consist of unstructured format capture from web. The repository is continuously challenged as Big data increases exponentially where security is one of the challenging and harmful concerns. To handle Big data, Hadoop technology takes cardinal part in analysis. I have proposed detection of DDos attack by using Counter based algorithm and Access Pattern algorithm which will implement using Map Reduce in Hadoop framework. Besides, we are predicting future behavior of attacker by means of analytical & statistical results.