Computing and supercomputing centers are usually provided with high-capacity wired network infrastructure. Wired technologies tend to be more stable and reliable, and interferences are mitigated by different layers. In addition, solutions such as optical fibers show a performance that exceeds any existing wireless technology. However, wired approaches also have important limitations. Although installation and maintenance are complex activities, whose cost increases exponentially with the size of the computing center, the resulting infrastructure is totally rigid and does not allow any flexible adaptation (although minimal). When the maximum capacity of wired links is reached, the probability of error and packet losses scale uncontrollably. And only an infrastructure enhancement can rectify the problem. But this scheme cannot adapt dynamically and may require months to be prepared and implemented. In the end, for most computing centers, traffic is considered stationary in the long term and no significant change is ever expected. However, in some transitory circumstances, computing centers may suffer an unexpected increase in traffic flows. For example, in response to any society event or technological problem. In those scenarios, the network capacity is desired to increase dynamically and temporarily, as the increase in demand is not sustainable. So, technological solutions to enable this dynamic adaptation are required. Therefore, in this paper we propose a resource sharing scheme, where servers in computing centers may request some temporal increased resources through a negotiation process supported by Blockchain networks and 5G technologies. Blockchain nodes will analyze the request to ensure no abuse is allowed and employ a Particle Swarm Optimization algorithm to decide which resource distribution has the biggest impact in global capacity. The charging process uses a mathematical model to calculate fees. Any dynamic resource provision is based on 5G mobile communications, whose behavior is modeled using common electromagnetic laws. An experimental validation based on simulation tools is also provided. Results show that the proposed solution can satisfy all temporary traffic increments, with up to 91% probability.