The costs of large computing facilities like data centers were dominated by the costs of the information technology (IT) equipment that they have in 1990s. As the indirect cost of IT has increased, Total Cost of Ownership(TCO) analysis is required to find the lifetime costs of acquiring, operating, and changing IT equipment. Nowadays, electricity-related costs such as the electrical power used by IT equipment and the facility costs associated with powering and cooling IT equipment has sharply grown. Nonetheless, those electricity-related costs have been underestimated up to now. In this paper, we will develop TCO model for a high performance computing data center and perform TCO analysis. We will also show that the trend towards growing electricity-related IT equipment costs continue, direct IT equipment acquisition costs will not be an critical factor of the economics of computing services.

The universe is thought to be filled with not only Standard Model (SM) matters but also dark matters. Dark matter is thought to play a major role in its construction. However, the identity of dark matter is as yet unknown, with various search methods from astrophysical observartion to particle collider experiments. Because of the cross-section that is a thousand times smaller than SM particles, dark matter research requires a large amount of data processing. Therefore, optimization and parallelization in High Performance Computing is required. Dark matter in hypothetical hidden sector is though to be connected to dark photons which carries forces similar to photons in electromagnetism. In the recent analysis, it was studied using the decays of a dark photon at collider experiments. Based on this, we studies double dark photon decays at lepton colliders. The signal channels are e+e– → AʹAʹ and e+e– → AʹAʹγ where dark photon Aʹ decays dimuon. These signal channels are based on the theory that dark photons only decay into heavily charged leptons, which can explain the muon magnetic momentum anomaly. We scanned the cross-section according to the dark photon mass in experiments. MadGraph5 was used to generate events based on a simplified model. Additionally, to get the maximum expected number of events for the double dark photon channel, the detector efficiency for several center of mass (CM) energy were studied using Delphes and MadAnalysis5 for performance comparison. The results of this study will contribute to the search for double dark photon channels at lepton colliders.