Zichen Xu

Exploring power-performance tradeoffs in database systems

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.1

PET: reducing database energy cost via query optimization

Energy conservation is a growing important issue in designing modern database management system (DBMS). This requires a deep thinking about the tradeoffs between energy and performance. Despite the significant amount of efforts at the hardware level to make the major components consume less energy, we argue for a revisit of the DBMS query processing mechanism to identify and harvest the potential of energy saving. However, the state-of-art architecture of DBMS does not take energy usage into consideration in its design. A major challenge in developing an energy-aware DBMS is to design and implement a cost-based query optimizer that evaluates query plans by both performance and energy costs. By following such a strategy, our previous work revealed the fact that energy-efficient query plans do not necessarily have the shortest processing time. This demo proposal introduces PET -- an energy-aware query optimization framework that is built as a part of the PostgreSQL kernel. PET, via its power cost estimation module and plan evaluation model, enables the database system to run under a DBA-specified energy/performance tradeoff level. PET contains a power cost estimator that can accurately estimate the power cost of query plans at compile time, and a query evaluation engine that the DBA could configure key PET parameters towards the desired tradeoff. The software to be demonstrated will also include workload engine for producing large quantities of queries and data sets. Our demonstration will show how PET functions via a comprehensive set of views from its graphical user interface named PET Viewer. Through such interfaces, a user can achieve a good understanding of the energy-related query optimization and cost-based plan generation. Users are also allowed to interact with PET to experience the different energy/performance tradeoffs by changing PET and workload parameters at query runtime.

A system for energy-efficient data management

Energy consumption of computer systems has increased at a steep rate in recent years. Following extensive energyrelated research and practice in the hardware and OS communities, much attention has been paid to developing energy-efficient applications. With database systems being a heavy energy consumer in modern data centers, we face the challenge of designing DBMSs with energy as a first-class performance goal. This paper presents our on-goingwork in designing and implementing a DBMS that enables significant energy conservations while maintaining other performance targets. We follow two new strategies in DBMS implementation to achieve our system design goal. The first one is to change the resource consumption patterns via energy-aware query optimization and reorganizing data records to enable load consolidation in disks. The second strategy is active control of power modes of hardware (i.e., CPU and hard disks) toward energy reduction. Specifically, we use control-theoretic techniques to allowdynamic adjustment of CPU frequency and online data migration to achieve disk load consolidation. Preliminary results have shown the effectiveness of our design.

Dynamic Energy Estimation of Query Plans in Database Systems

Data centers are well known to consume large amounts of energy. Since database is one of the major applications in a typical data center, building energy-aware database systems has become an active research topic recently. The quantification of the energy cost of database systems is an important task in designing such systems. In this paper, we report our recent efforts on this topic, with a focus on the energy cost estimation of query plans during query optimization. We start from building a series of physical models for energy estimation of individual relational operators based on their resource consumption patterns. Since the execution of individual queries is a combination of relational operators, we use the physical models as a basis for a comprehensive energy cost estimation model for entire query plans. To further improve model accuracy under system dynamics and the variations of workload characteristics, we develop an online model estimation scheme that dynamically corrects the static model based on advanced modeling techniques adopted from control engineering. The models are implemented in a real database and evaluated on a physical test bed with a comprehensive set of experimental workloads. The results show that our solution achieves a high accuracy (above 90%) in energy estimation despite noises from the system and workloads.

Blending on-demand and spot instances to lower costs for in-memory storage

In cloud computing, workloads that lease instances on demand get to execute exclusively for a set time. In contrast, workloads that lease spot instances execute until a competing workload outbids the current lease. Spot instances cost less than on-demand instances, but few workloads can use spot instances because of the variable leasing period. We present BOSS, a framework that uses spot instances to reduce costs for in-memory storage workloads. BOSS uses on-demand instances to create and update objects. It uses spot instances to handle read-only queries. BOSS leases instances from multiple sites and exploits varying prices between the sites. When spot instances stop abruptly at one site, BOSS places newly created objects at other sites, reducing the impact on response time. BOSS proposes a novel, online replication approach (1) avoids placing data at too many sites and (2) provides O(1.5)-competitive ratio under skewed cost distributions. Within a site, BOSS manages the tradeoff between savings and risks from replicating to spot instances. We implemented BOSS on top of Cassandra and deployed it on up to 78 instances across 8 sites in Amazon and Google clouds. With BOSS hosting TPC-W data, we spent

Building a power-aware database management system

8 per hour on Amazon. For the same service, we spent

CADRE: Carbon-Aware Data Replication for Geo-Diverse Services

55 per hour to use ElastiCache and

Online Energy Estimation of Relational Operations in Database Systems

49 per hour to use on-demand instances only. BOSS saved 85% and 84% respectively. Further, BOSS achieved 95th percentile response time within 13% of ElastiCache.

Tell-tale tails: Decomposing response times for live internet services

In todays large-scale data centers, energy costs (i.e., the electricity bill) are projected to outgrow that of hardware. Despite a long history of research in energy-saving techniques, especially low-power hardware, little work has been done to improve the power efficiency of data management software. Power-aware computing research at the application level has been found to be synergistic to that at the hardware and OS levels because it can provide more opportunities for energy reduction in the underlying systems. This paper describes the authors thesis work on creating a power-aware database management (P-DBMS) and initial ideas on the design of such systems, with the focus on a power-aware query optimization module inside the DBMS. We discuss the main technical challenges in designing the optimizer and present our strategies to meet such challenges. We focus our discussions on a power model to accurately measure the energy costs of query executions plans, and a cost evaluation model for plan selection. An important feature of this work is the formal control-theoretic methods we use to model and optimize the database towards the performance and energy saving goals. This rigorous design methodology is in sharp contrast to heuristic-based adaptive solutions that rely on extensive empirical evaluation and manual tuning. Our experiments using a power-aware query optimizer under our initial design show that there exist significant potential in power/energy savings.

Power-aware DBMS: potential and challenges

Internet services replicate data to geo-diverse sites around the world, often via consistent hashing. Collectively, these sites span multiple power authorities that independently control carbon emissions at each site. Serving data from a carbon-heavy site increases the services carbon footprint, but it is hard to place data at sites that will have low emission rates without replicating to too many sites. We present CADRE, a carbon-aware data replication approach. CADRE forecasts emission rates at each site and replicates data to sites that combine together to yield low carbon footprints. It makes replication decisions online, i.e., When data is created, and thus avoids emissions caused by moving data frequently in response to changing emission rates. CADRE uses the multiple-choice secretary algorithm to replicate objects with large footprints to low emission sites. It models carbon footprints for each object using the footprint-replication curve, a graph that maps replication factors to expected carbon footprints. CADRE also achieves availability goals, respects storage capacity limits and balances data across sites. Compared to consistent hashing, our approach reduces carbon footprints by 70%. It also supports and enhances the state-of-the-art green load balancing, reducing the carbon footprint by an additional 21%.