CSE 691: Energy-Efficient Computing Lecture 7 SMARTS: custom-made systems Anshul Gandhi 1307, CS building

memcache

Benchmark competitions

fawn paper

Data-intensive workloads Seek-bound small random examples? problems? Scan-bound large sequential examples? problems?

Why FAWN? 1.Memory wall (??) 2.Increased CPU power consumption 3.DVFS is limited Modern CPUs operate close to V min Constant leakage current 4.Peak power and data center density

FAWN results

Processor scaling trends 1.Moore’s law (observation) # transistors/chip ↑ 2X/2yr (how?) Frequency ↑ as transistor size ↓ (max 9GHz) Leakage current/power ↑ as transistor size ↓ Heat ↑ as frequency ↑

Processor scaling trends 2.Dennard scaling (observation) Transistor power (V+I) ↓ as transistor size ↓ + Moore’s law = perf/watt ↑ 2X/2yr (how?) Not true now due to leakage current So we did multicore!

heteromates paper

Dark silicon Cannot power on all of the CPU Result of: 1.Success of Moore’s law 2.Failure of Dennard scaling Turbo boost

Main ideas Heterogeneous cores 1.For different performance requirements 2.Energy considerations (battery vs. plugged) 3.Thermal considerations 4.Dark silicon

Main ideas