Compute load shaping using virtual capacity and preferential location real time scheduling

A system comprising:

• one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
• for each cell of a plurality of cells, each cell of the plurality of cells being at different respective geographic locations:
• obtaining a load forecast for a future period of time that indicates forecasted future compute load for the cell during respective intervals of the future period of time;
• obtaining a power model that models a relationship between power usage and computational usage for the cell;
• and
• obtaining a carbon intensity forecast that indicates a forecast of carbon intensity for the geographic area where the cell is located for the respective intervals of the future period of time;
• determining, from a first carbon intensity forecast for a first cell of the plurality of cells and a second carbon intensity forecast for a second cell of the plurality of cells, that the carbon intensity for one or more time intervals for the first geographic area where the first cell is located is less than the carbon intensity for one or more time intervals for the second geographic area where the second cell is located;
• and
• in response to the determination, shifting a compute load from the second cell to the first cell during at least one of the one or more time intervals.

A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:

• for each cell of a plurality of cells, each cell of the plurality of cells being at different respective geographic locations:
• obtaining a load forecast for a future period of time that indicates forecasted future compute load for the cell during respective intervals of the future period of time;
• obtaining a power model that models a relationship between power usage and computational usage for the cell;
• and
• obtaining a carbon intensity forecast that indicates a forecast of carbon intensity for the geographic area where the cell is located for the respective intervals of the future period of time;
• determining, from a first carbon intensity forecast for a first cell of the plurality of cells and a second carbon intensity forecast for a second cell of the plurality of cells, that the carbon intensity for one or more time intervals for the first geographic area where the first cell is located is less than the carbon intensity for one or more time intervals for the second geographic area where the second cell is located;
• and
• in response to the determination, shifting a compute load from the second cell to the first cell during at least one of the one or more time intervals.