Interactive　network　scenario　reproduction　is　an　important　method　for　generating　realistic　and　responsive　traffic　workloads.　However,　due　to　the　complex　traffic　generation　mechanisms,　controlling　traffic　volume　based　on　accurate　mathematical　models　in　interactive　network　scenario　reproduction　becomes　a　challenging　problem.　In　this　paper,　we　design　and　implement　a　volume　control　system　for　interactive　network　traffic　replay,　and　formulate　the　volume　control　task　as　a　target　tracking　problem　where　the　output　traffic　volume　is　regulated　through　adjustment　of　input　traffic　volume.　We　then　adopt　a　function　approximator　(FA)　in　our　system　to　generate　the　desired　input　volumes.　The　FA　characterizes　the　mapping　that　takes　system　measurements　as　input　and　directly　produces　controls　as　output.　Meanwhile,　simultaneous　perturbation　stochastic　approximation　(SPSA)　algorithm　is　employed　to　adaptively　estimate　the　parameters　in　control.　To　validate　the　system　and　the　method　we　have　implemented　it　and　conducted　experiments　with　actual　traffic　traces.　Empirical　studies　show　that　our　system　can　track　target　output　volumes　effectively　under　a　wide　range　of　network　conditions.　The　performance　of　the　control　system　is　further　investigated　in　terms　of　its　convergence　time,　generated　input/output　traffic　volume　and　target　tracking　error.