Ever increasing amount of data in networks may lead to congested networks, routers with full queues, increase of loss rate, and end-to-end latency. As a result effective throughput in networks would be plummeted. To deal with such challenges routers have to route packets faster and more precise. This concern seems to be more important in the backbones and regions near datacenters (which are crucial segments of nowadays networks), since they must tolerate with tremendous amount of traffic. Researchers have proposed different routing lookup algorithms, but since most of those need some memory accesses, routing process speed is limited to memory speed. In this paper, concentration is on flooding traffic beyond datacenters. Beyond doubt decreasing traffic volume in the regions close to datacenters, which absorb large proportion of traffic, can effectively improve congestion status and overall throughput of network. This research proposes that network can be segmented into regions according to distance of nodes from datacenters and backbones and different policies and routing algorithms would function in each segment. This method can effectively deduct the load on hardware and software in networks. Results show notable improvements in Round Trip Time (RTT), latency, and queue length parameters in routers.

Datacenter Congestion control Cooperative caching