1 / 22

Towards Eco-friendly Database Management Systems W. Lang, J. M. Patel (U Wisconsin), CIDR 2009

Towards Eco-friendly Database Management Systems W. Lang, J. M. Patel (U Wisconsin), CIDR 2009. Shimin Chen Big Data Reading Group. Introduction. Energy consumption is important for data centers: 2005: 1.2% of total US energy consumption is attributed to powering and cooling servers, ~ $2.7B

temple
Télécharger la présentation

Towards Eco-friendly Database Management Systems W. Lang, J. M. Patel (U Wisconsin), CIDR 2009

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Towards Eco-friendly Database Management SystemsW. Lang, J. M. Patel (U Wisconsin), CIDR 2009 Shimin Chen Big Data Reading Group

  2. Introduction • Energy consumption is important for data centers: • 2005: 1.2% of total US energy consumption is attributed to powering and cooling servers, ~ $2.7B • If current methods for powering data centers continue, the consumption will nearly double by 2011 • For DBMS: • Previously large ignored energy efficiency • Must start considering energy as a critical metric

  3. This paper: ecoDB • New project: energy efficient data processing techniques • Two broad classes of techniques: • “global”: change how entire system is managed or used • E.g. job scheduling • “local”: improve methods of processing data at individual nodes (focus of the paper)

  4. Idea

  5. Two Questions • (1) “How does a system generate graphs as shown in Figure 1?” • DMBS must know HW capabilities and operating characteristics • Accurately estimate / continuously measure energy consumption • (2) “How can such a graph be used?” • Systematic method to change settings • Service level agreements (SLAs) • This paper focuses on mechanisms for creating graphs

  6. Outline • Introduction • Processor Voltage/Frequency Control (PVC) • Improved Query Energy-efficiency by Introducing Explicit Delays (QED) • Opportunities for Energy Efficiency • Summary

  7. Techniques • CPU freq = front side bus (FSB) freq * CPU multiplier • DVFS (dynamic voltage and frequency scaling) • Each p-state defines a CPU multiplier • CPU voltage is based on CPU multiplier • Under-clocking (Focus of this paper) • Reduce FSB freq • Finer granularity • Also changes RAM freq

  8. System Under Test • System components: • ASUS P5Q3 Deluxe Wifi-AP motherboard • Intel Core2-Duo E8500 • 2×1GKingston DDR3 main memory • ASUS GeForce 8400GS 256M • Western Digital Caviar SE16 320G SATA disk • Power supply unit (PSU): a Corsair VX450W PSU • System power draw measured by a Yokogawa WT210 unit (suggested by SPEC Power benchmark) • MS Windows Server 2008 • JDBC (Java 1.6)

  9. Power • CPU power sensors on motherboard: • ASUS motherboard has an EPU processor that directly measures the CPU power. • ASUS P5Q3 Deluxe 6-Engine software displays information gathered from this hardware sensor. • Current CPU wattage displayed in GUI: • The authors sample the GUI every second • Compute CPU joules using the average CPU wattage and the execution time of a workload

  10. Component powers • No hard disk, no operating system • Focusing on CPU power: • CPU power consumption is often about 25% of the total system power consumption in the experiments

  11. DB test • Workload • Use a commercial DBMS and MySQL 5.1.28 • TPC-H (ad-hoc decision support), scale factor 1.0 (1GB data) • Only run Query 5: six table join and a group by • A run consists of ten queries with different parameters • FSB underclocking (allowed by ASUS 6-engine software) • Stock (normal) • Reduce FSB freq by 5%, 10%, and 15% • CPU voltage downgrade • “small” and “medium” downgrade • 7 settings: • Stock, 3 FSB freq reductions X 2 CPU voltage downgrades

  12. Equal Energy delay product

  13. Equal Energy delay product With the same voltage level, larger frequency the better EDP

  14. Theoretical Modeling • EDP= joules x times = power x time2= power / freq2 • Power=CV2F • EDP = CV2/F

  15. Disk Energy • Measured separately for stock setting • Warm database • CPU: 1228.7 Joules • Disk: 214.7 Joules • Cold database • CPU: 2146.0 Joules • Disk: 1135.4 Joules

  16. Outline • Introduction • Processor Voltage/Frequency Control (PVC) • Improved Query Energy-efficiency by Introducing Explicit Delays (QED) • Opportunities for Energy Efficiency • Summary

  17. Idea • Explicitly delay queries • look for commonalities among multiple queries • Group multiple queries into a single query • After execution, split query results

  18. Setting • DB clients repeatedly issue single table select queries with different selection predicate. • For example: SELECT *FROM lineitemWHERE l_quantity=X • DBMS processes one query at a time • QED: buffer queries in a queue, merge them, send the merged query, split results • In the experiments, X is different for the queries, so no overlaps

  19. As batch size increases, diminishing decrease in energy consumption.

  20. Outline • Introduction • Processor Voltage/Frequency Control (PVC) • Improved Query Energy-efficiency by Introducing Explicit Delays (QED) • Opportunities for Energy Efficiency • Summary

  21. Opportunities in (DBMS) Software • Traditional DB investigations into improving query response times • Energy vs. performance tradeoffs • Operator-level: rethink join algorithms • Query-level: energy-efficient query plans • Workload management per server • Workload management for the entire collection of servers: scheduling and using techniques to turn entire servers off

  22. Summary • Energy-efficient data processing • Studied two techniques • Processor Voltage/Frequency Control (PVC) • Improved Query Energy-efficiency by Introducing Explicit Delays (QED) • Designing a DBMS to balance the response time vs. energy consumption opens a wide range of research issues

More Related