It is common for cloud data centers meeting unexpected loads like requestbursts, which may lead to overloaded situation and performance degradation.Dynamic Voltage Frequency Scaling and VM consolidation have been provedeffective to manage overloads. However, they cannot function when the wholedata center is overloaded. Brownout provides a promising direction to avoidoverloads through configuring applications to temporarily degrade userexperience. Additionally, brownout can also be applied to reduce data centerenergy consumption. As a complementary option for Dynamic Voltage FrequencyScaling and VM consolidation, our combined brownout approach reduces energyconsumption through selectively and dynamically deactivating applicationoptional components, which can also be applied to self-contained microservices.The results show that our approach can save more than 20% energy consumptionand there are trade-offs between energy saving and discount offered to users.
展开▼
