Petaflop in a cubic meter

HPCWire – Last year, John Kelly from IBM suggested that a byproduct of success in building an exaflop computer would be a petaflop in 1/3 of a rack. If we assume that DOE’s target of 20 megawatts (MW) power consumption for an exascale system is achieved and that 1/3 of a rack is about one cubic meter, then a petaflop in a cubic meter box would consume about 20 kilowatts (kW).

Such a system would consume about as much power as 4 electric clothes dryers. If we wanted to purchase a dedicated off-grid power supply for a petaflop box, we could find one on the internet for about $5,000. (Then we could measure flops/gallon!) On the US electric grid, the average price of 1 kWh in 2011 was 11.20 cents. So, one could operate the system continuously for a year at a power cost of about $20,000. These may be oversimplifications, but you get the point.

Currently, one Blue Gene/Q cabinet has a volume of just over three cubic meters, holds hardware with a theoretical peak performance of just over 200 teraflops, and typically consumes about 65 kW. So using this technology as an example, to get a petaflop in a cubic meter we’d need to reduce the volume and power consumption by a factor of three and increase performance by a factor of five.

In 2011 the initial IBM pricing for a petaflop was $150 million.

A balanced configuration, with eight Power 775 nodes and two disk nodes, will run you about $8.1m per rack and deliver 64 teraflops of raw computing oomph. Scale that up to 1,365 compute nodes and 342 storage nodes – assuming the workload needs a reasonable amount of local disk – and you are at 10.9 petaflops of raw performance, 2.7PB of memory, and 26.3PB of disk/flash storage. That will also run you something around $1.5bn at list price.

If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks

About The Author