Abstract:Modern processors often need to switch among different power states based on usage scenarios, energy availability, and thermal conditions. Dynamic voltage and frequency scaling (DVFS) is a commonly used power management strategy to trade off performance and energy. As transistor scaling is reaching its limit, the viable supply voltage range where DVFS can operate is shrinking, which limits its effectiveness. To extend the performance-energy trade-off capabilities in modern processors, this article proposes dynamic core scaling (DCS) that does not rely on voltage scaling. DCS dynamically adjusts the active superscalar datapath resources so that programs run at a given percentage of their maximum speed while minimizing energy consumption at the same time. Since DCS does not need voltage scaling, it can be combined with DVFS to achieve greater energy savings. To effectively manage performance-energy trade-offs using a combination of DCS and DVFS, this article proposes an oracle controller that demonstrates the optimal control strategy, and two practical controllers that are applicable in real implementations. Evaluations using an 8-way superscalar processor implemented in a 45-nm circuit show that DCS is more effective in performance-energy trade-offs than DVFS at the high performance end for a number of SPEC CPU2000 benchmarks. When used together with DVFS, DCS saves an additional 20 percent of a full-size core's energy on average. At the minimum operating voltage, DVFS hits its limit, while DCS is still able to achieve an average of 46 percent further energy reduction.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is that in modern processors, as transistor sizes shrink, the effective voltage range of Dynamic Voltage and Frequency Scaling (DVFS) becomes narrower, thus limiting its ability to make trade - offs between performance and energy consumption. To overcome this limitation, the paper proposes a new technique - Dynamic Core Scaling (DCS), which does not rely on voltage scaling but achieves the trade - off between performance and energy consumption by dynamically adjusting the effective superscalar data - path resources in large out - of - order execution cores. Specifically, DCS can dynamically adjust the front - end width, dispatch width, and the sizes of the Instruction Queue (IQ), Load/Store Queue (LSQ), and Reorder Buffer (ROB), enabling the program to run at a certain percentage of the maximum speed while minimizing energy consumption. In addition, DCS can be used in combination with DVFS to achieve greater energy - saving effects.
The main contributions of the paper include:
- Proposing DCS as an orthogonal alternative to DVFS, achieving the trade - off between performance and energy consumption by dynamically adjusting superscalar data - path resources rather than voltage.
- Designing a hardware DCS controller that can minimize energy consumption while ensuring that the application runs at a given percentage of the maximum speed.
- Proposing three control mechanisms (including an ideal controller, a simple controller, and a complex controller) to effectively manage the trade - off between performance and energy consumption using the combination of DCS and DVFS, and providing a detailed analysis of these control mechanisms in different scenarios.
- Verifying the effectiveness of DCS and its combination with DVFS through experiments and comparing it with using DVFS alone.
- Evaluating the overhead of DCS and the resulting energy - saving effects, using a 45 - nanometer circuit infrastructure.
Through these contributions, the paper aims to expand the ability of modern processors to make trade - offs between performance and energy consumption at the high - performance end, especially when DVFS reaches its limit, DCS can still achieve further energy savings.