# Power-Performance Tradeoffs in Data Center Servers: DVFS, CPU pinning, Horizontal, and Vertical Scaling

## TL;DR

This paper evaluates four power-performance control techniques in data center servers, quantifying their effects on response time, throughput, and power consumption through experiments with Wikipedia replicas.

## Contribution

It provides a comprehensive experimental analysis of DVFS, CPU pinning, and scaling techniques, revealing their tradeoffs and practical impacts on server performance and energy use.

## Key findings

- DVFS rarely reduces underloaded server power by more than 5%.
- DVFS can limit saturated server power by up to 20% with performance cost.
- Horizontal and vertical scaling improve response times but not proportionally to resources added.

## Abstract

Dynamic Voltage and Frequency Scaling (DVFS), CPU pinning, horizontal, and vertical scaling, are four techniques that have been proposed as actuators to control the performance and energy consumption on data center servers. This work investigates the utility of these four actuators, and quantifies the power-performance tradeoffs associated with them. Using replicas of the German Wikipedia running on our local testbed, we perform a set of experiments to quantify the influence of DVFS, vertical and horizontal scaling, and CPU pinning on end-to-end response time (average and tail), throughput, and power consumption with different workloads. Results of the experiments show that DVFS rarely reduces the power consumption of underloaded servers by more than 5%, but it can be used to limit the maximal power consumption of a saturated server by up to 20% (at a cost of performance degradation). CPU pinning reduces the power consumption of underloaded server (by up to 7%) at the cost of performance degradation, which can be limited by choosing an appropriate CPU pinning scheme. Horizontal and vertical scaling improves both the average and tail response time, but the improvement is not proportional to the amount of resources added. The load balancing strategy has a big impact on the tail response time of horizontally scaled applications.