The seismic exploration is primary tool in search for oil and gas exploration. The results from these explorations are used to produce images of the earth's subsurface, that geologists analyze for understanding the earth subsurface picture. There are several processing techniques involved in processing data from seismic explorations. Seismic migration process is one of the most computationally intensive steps of all the seismic data processing sequence. Migration techniques are highly compute and I/O intensive and therefore require high performance systems to carry out the operations efficiently. It is well evident that rather than sequential machines parallel computers provide cost-effective solutions as migration algorithms show lot of potential for parallelism. One such algorithm is Kirchhoff migration. The processing of Kirchhoff migration algorithm involves handling large volumes of data which needs high computational power. Most of the exploration companies cannot afford to have computational power or the estimate of the power they need on their own. Therefore there should be a model or technique by which oil and gas exploration companies can find out an estimate for the computational power they need. This thesis gives a model for estimating the computer power needed to run Kirchhoff migration algorithm for a given volume of data.