There are two generally accepted laws taken into account when talking about improvements in the technology world. On one hand, Moore's law, named after Intel's co-founder, provides that the number of transistors in a dense integrated circuit doubles about every two years (with a distinct exception that, since 2010, it's been closer to every three years). On the other hand, Kryder's law, mentioned in 2005 by Seagate's chief technology officer, postulates that disk storage space increases even more than what Moore had accounted for. Technology may advance at a fast rate, but we are still far from being able to gather all required computing power for large-scale projects without needing equivalent large amounts of money, and this can prove quite stringent in life sciences research, aerodynamics or other R&D areas.
UberCloud has been trying to change this for the past couple of years, acting as a marketplace for computing-as-a-service resources, providing access to large computers tuned for engineering and research at professionally managed and secured data centers. This all leads to a better, faster and more reliable workflow during R&D. You can submit projects to obtain quotes for service providers of computational force, and you can also join other projects already unveiling. Providers included on the platform are software, resource, middleware & tools and expertise & support providers.
One of the most recent announcements came at the beginning of February, UberCloud announced the availability of high performance containers build on top of the Docker technology. The containers are portable allowing simple, transparent on-demand access to applications and data without major impact over appearance of the workstation. The way that the marketplace works is buying out different time packages for CPU core environment, and this could lead to getting ahead on R&D big time.