Abstract
A workload-aware low-power neuromorphic controller for dynamic voltage and frequency scaling (DVFS) in very large scale integration (VLSI)systems is presented. The neuromorphic controller predicts future workload values and preemptively regulates supply voltage and frequency based on past workload profile. Our specific contributions include: 1) implementation of a digital and analog version of the controller in 45 nm CMOS technology, resulting in a 3% performance hit with a power overhead in the range of 10-150 μW from the controller circuit; 2) higher prediction accuracy compared to a software based OS-governed DVFS scheme, reducing wasted power and improving error margins; and 3) power savings of up to 52% and improvement of up to 15% compared to the OS-based scheme. The digital design has minimal power overhead and is more reconfigurable, while analog design is better suited for nonlinear and complex computational tasks.
| Original language | English (US) |
|---|---|
| Article number | 6035994 |
| Pages (from-to) | 381-390 |
| Number of pages | 10 |
| Journal | IEEE Journal on Emerging and Selected Topics in Circuits and Systems |
| Volume | 1 |
| Issue number | 3 |
| DOIs | |
| State | Published - Sep 2011 |
| Externally published | Yes |
Bibliographical note
Funding Information:Dynamic voltage and frequency scaling (DVFS) is supported by nearly all commercially available general purpose microprocessors under trademarks such as Intel’s Enhanced Speedstep Technology and AMD’s Cool’n’Quiet and PowerNow! technologies. These technologies work in using an open standard specification known as Advanced Configuration and Power Interface (ACPI)1 developed for Operating System-directed configuration and Power Management (OSPM).
Keywords
- Adaptive
- neuromorphic
- power management
- simulation
- very large scale integration (VLSI)