Not many would argue that humankind has an effect on the environment, for example CO2 concentrations now regularly exceeding 400 ppm. IT is one of the fastest growing contributors to rising CO2 levels due to its energy consumption. IT also has significant other environmental footprints (requirements for exotic minerals, water for cleansing during manufacture, and disposal issues).
High end compute, data centres for "Big Data", "Machine Learning", "Deep Learning" and other contemporary paradigms, and networking all play a major part in Greener Compute. And Greener Compute is more than the energy to run a system, to us it includes embodied emissions as well as the environmental impact of materials used in construction, life cycle and disposal.
But there are a myriad of ways to tackle the growing trend of energy consumption as we highlight in our recent presentation. (Commentary to follow)
High End Compute provides consultancy services on #greenerCompute and acts as a informal network of like-minded R&D:
- Who's who from Adept to Oxford
- Some high level approaches to implementing Greener Compute
- Discussion on what to measure and ideas on how to do so
- our blog on determining the power/performance footprint of a Jetson for linear algebra (in draft [June2015])
- please get in touch if you want to discuss improving PUE, improving code efficiency or switching for #greenerCompute