Simulation of a Big data streams data center (BDSDC) reengineered for efficient dumpsite datastream offloading. A use case municipal waste management aggregation data center network (DCN) and cloud-driven micro-services orchestration at the edge with low latency translation into the cloud environments. Discrete-Event Modelling and Simulation Methodology (DEMSM) was adopted. Using the experimental test data gathered from the experimental testbed (UNN DCN), a simulation study was carried out in Riverbed Modeller while allowing for result comparison with the trace file of the typical traditional DCN. It was discovered that BDMSC performed much better that the traditional DCN and addressed majority of the challenges.  The results of the proposed BDSDC system considered BDSCA Optimization, and non-BDSCA Optimization use-cases. Second, the proposed BDSA was then compared with Bayesian and MapReduce algorithms. With BDSCA Optimization, 47.37% data stream workload is provisioned which is very useful in deterministic traffic workloads. This is in contrast with the best-efforts scheme that yielded 52.63%. In terms of throughput, proposed BDSA offered 52.63% throughput cycles while Bayesian and MapReduce gave 36.84% and 10.53% each. Considering network latency, the proposed BDSCA latency optimization is shown to be very attractive at 27.77%. This is certainly better than Bayesian (55.56%) and MapReduce (16.67%). In terms of resource utilization, at peak traffic, all the algorithms had similar trend pattern even at the steady and relaxed states. At a closer experimental control and monitoring, the resource utilization, MapReduce Apriori, Bayesian and the proposed BDSCA offered 37.55%, 25.03% and 37.42% respectively. Finally, this is better than reactive DCell and BCube integration cloud domains.