Abstract-With the total energy consumption of computing systems increasing in a steep rate, much attention has been paid to the design of energy-efficient computing systems and applications. So far, database system design has focused on improving performance of query processing. The objective of this study is to experimentally explore the potential of power conservation in relational database management systems. We hypothesize that, by modifying the query optimizer in a DBMS to take the power cost of query plans into consideration, we will be able to reduce the power usage of database servers and control the tradeoffs between power consumption and system performance. We also identify the sources of such savings by investigating the resource consumption features during query processing in DBMSs. To that end, we provide an in-depth anatomy and qualitatively analyze the power profile of typical queries in the TPC benchmarks. We perform extensive experiments on a physical testbed based on the PostgreSQL system using workloads generated from the TPC benchmarks. Our hypothesis is supported by such experimental results: power savings in the range of 11% -22% can be achieved by equipping the DBMS with a query optimizer that selects query plans based on both estimated processing time and power requirements.

No abstract

Abstract-Entering the era of cloud computing, data centers are scaling in a fast pace. However, as the increasing number of servers being deployed in data centers, the data center power distribution systems have already approached peak capacities. Since the upgrades of the power systems are extremely expensive, power oversubscription has become a trend in modern data centers as a cost-effective way to handle power provisioning. Under benign workload of data centers, power oversubscription works well as servers rarely peak simultaneously. However, power oversubscription makes data centers vulnerable to malicious workload that can generate power spikes on multiple servers at the same time, which may cause branch circuit breakers to trip and lead to undesired power outages. In this paper, we introduce a new security concept called power attack and exploit the attack vectors in platform as a service (PaaS), infrastructure as a service (IaaS), and software as a service (SaaS) cloud environments, respectively. To demonstrate the feasibility of launching a power attack, we conduct series of hardware experiments and datacenter-level simulations. Moreover, we give a detailed analysis on how different power management methods can affect a power attack and how to mitigate such an attack. Our experimental results and analysis show that power attacks will pose a serious threat to modern data centers and should be taken into account while deploying new high-density servers and power management techniques.

No abstract

No abstract