Power Partitioning for Multiprocessors Kai Ma, Amiri-Kamalabad Mojtaba.

Slides:



Advertisements
Similar presentations
Application-Aware Memory Channel Partitioning † Sai Prashanth Muralidhara § Lavanya Subramanian † † Onur Mutlu † Mahmut Kandemir § ‡ Thomas Moscibroda.
Advertisements

March 2009 Emissions Trading in South Africa National Climate Change Summit Emily Tyler.
A Cyber-Physical Systems Approach to Energy Management in Data Centers Presented by Chen He Adopted form the paper authors.
Adaptive Routing in (Q)NoC
John W. Gardner Center for Youth and Their Communities DRAFT Youth Development Overview, 2009 DRAFT.
Enhancing the Platform Independence of the Real-Time Specification for Java Andy Wellings, Yang Chang and Tom Richardson University of York.
1 Energy-efficiency potential of a phase-based cache resizing scheme for embedded systems G. Pokam and F. Bodin.
Akhil Langer, Harshit Dokania, Laxmikant Kale, Udatta Palekar* Parallel Programming Laboratory Department of Computer Science University of Illinois at.
University of Michigan Electrical Engineering and Computer Science University of Michigan Electrical Engineering and Computer Science 1 Maestro: Orchestrating.
Integrated Regulation for Energy- Efficient Digital Circuits Elad Alon 1 and Mark Horowitz 2 1 UC Berkeley 2 Stanford University.
Process: A Generic View n A software process  is a roadmap to building high quality software products.  provides a framework for managing activities.
PARAID: The Gear-Shifting Power-Aware RAID Charles Weddle, Mathew Oldham, An-I Andy Wang – Florida State University Peter Reiher – University of California,
Tufts Wireless Laboratory School Of Engineering Tufts University “Network QoS Management in Cyber-Physical Systems” Nicole Ng 9/16/20151 by Feng Xia, Longhua.
Common Carbon Metric for Measuring Energy Use & Reporting Greenhouse Gas Emissions from Building Operations A tool developed by GHG Protocol and UNEP-SBCI.
A Framework for Distributed Model Predictive Control
COMP 2923 Basics of Green IT Danny Silver JSOCS, Acadia University.
Ramazan Bitirgen, Engin Ipek and Jose F.Martinez MICRO’08 Presented by PAK,EUNJI Coordinated Management of Multiple Interacting Resources in Chip Multiprocessors.
Adaptive Power Shifting in Server Systems Ming Chen Xue Li.
1 Customer-Aware Task Allocation and Scheduling for Multi-Mode MPSoCs Lin Huang, Rong Ye and Qiang Xu CHhk REliable computing laboratory (CURE) The Chinese.
1 The Success of Mergers & Acquisitions in Relation to the Alignment of the Acquirer and the Target Company’s Strategic Aggressiveness and Capability Responsiveness.
Applying Control Theory to the Caches of Multiprocessors Department of EECS University of Tennessee, Knoxville Kai Ma.
1 Tuning Garbage Collection in an Embedded Java Environment G. Chen, R. Shetty, M. Kandemir, N. Vijaykrishnan, M. J. Irwin Microsystems Design Lab The.
Background Gaussian Elimination Fault Tolerance Single or multiple core failures: Single or multiple core additions: Simultaneous core failures and additions:
Reconfigurable Computing Using Content Addressable Memory (CAM) for Improved Performance and Resource Usage Group Members: Anderson Raid Marie Beltrao.
Joint Power Optimization Through VM Placement and Flow Scheduling in Data Centers DAWEI LI, JIE WU (TEMPLE UNIVERISTY) ZHIYONG LIU, AND FA ZHANG (CHINESE.
CLIC Implementation Studies Ph. Lebrun & J. Osborne CERN CLIC Collaboration Meeting addressing the Work Packages CERN, 3-4 November 2011.
Towards Dynamic Green-Sizing for Database Servers Mustafa Korkmaz, Alexey Karyakin, Martin Karsten, Kenneth Salem University of Waterloo.
Overview What do we mean by a Learning Organisation? Why did we develop a People Development Framework? What was the process involved in building the.
Process: A Generic View
Optimal Power Allocation for Multiprogrammed Workloads on Single-chip Heterogeneous Processors Euijin Kwon 1,2 Jae Young Jang 2 Jae W. Lee 2 Nam Sung Kim.
1 Vertex-detector power pulsing CLIC Detector and Physics Collaboration Meeting, 02/10/2013 Cristian Alejandro Fuentes Rojas, PH-ESE-FE,
Paulius Baniūnas Ministry of Finance of the Republic of Lithuania EU Structural Support Management Department Monitoring and Analysis Division SYSTEM OF.
1 Performance elements in budget and reporting process - Norway 5TH ANNUAL MEETING OF OECD SENIOR BUDGET OFFICIALS NETWORK ON PERFORMANCE&RESULTS – 28.
Optimizing Power and Energy Lei Fan, Martyn Romanko.
1 Thermal Management of Datacenter Qinghui Tang. 2 Preliminaries What is data center What is thermal management Why does Intel Care Why Computer Science.
ECE555 Topic Presentation Energy-efficient real-time scheduling Xing Fu 20 September 2008 Acknowledge Dr. Jian-Jia Chen from ETH providing PPT Slides for.
Hardware Architectures for Power and Energy Adaptation Phillip Stanley-Marbell.
Computer Science and Engineering Power-Performance Considerations of Parallel Computing on Chip Multiprocessors Jian Li and Jose F. Martinez ACM Transactions.
Computing Systems: Next Call for Proposals Dr. Panagiotis Tsarchopoulos Computing Systems ICT Programme European Commission.
1 Developing Accountability in a Large Decentralized Agency U.S. Forest Service Raymond S. Thompson
The renewables Directive 1.Sets mandatory national targets for renewable energy shares, including 10% biofuels share, in 2020 (Articles 3 and 5) 2.Requires.
ECE555 Course Project Proposal Coordinated Power and Utilization Control for DRE System with E2E Tasks Xing Fu Eric Puster 16 September 2008.
Chapter 4 Motor Control Theories Concept: Theories about how we control coordinated movement differ in terms of the roles of central and environmental.
Thomas G. Cummings Christopher G. Worley Chapter Twenty : Transformational Change Organization Development and Change.
ECE692 Course Project Proposal Cache-aware power management for multi-core real-time systems Xing Fu Khairul Kabir 16 September 2009.
Best detection scheme achieves 100% hit detection with
Parapet Research Group, Princeton University EE Workshop on Hardware Performance Monitor Design and Functionality HPCA-11 Feb 13, 2005 Hardware Performance.
SizeCap: Efficiently Handling Power Surges for Fuel Cell Powered Data Centers Yang Li, Di Wang, Saugata Ghose, Jie Liu, Sriram Govindan, Sean James, Eric.
M AESTRO : Orchestrating Predictive Resource Management in Future Multicore Systems Sangyeun Cho, Socrates Demetriades Computer Science Department University.
Marilyn Wolf1 With contributions from:
Strategic Management (MGT501)
Start End What is a project? Definition from PMBOK -
Andrea Acquaviva, Luca Benini, Bruno Riccò
Xiaodong Wang, Shuang Chen, Jeff Setter,
Computing Resource Allocation and Scheduling in A Data Center
Évora Demonstrator Site
Houssam-Eddine Zahaf, Giuseppe Lipari, Luca Abeni RTNS’17
Unistore: Project Updates
Some challenges in heterogeneous multi-core systems
Hui Chen, Shinan Wang and Weisong Shi Wayne State University
Digital Processing Platform
Tosiron Adegbija and Ann Gordon-Ross+
TRANSFORMATIONAL CHANGE
Organization Development and Change
Organization Development and Change
Developing Your KPIs and Measuring Your AFT
Tosiron Adegbija and Ann Gordon-Ross+
Industry Engagement Program Medical Diagnostic Imaging (MDI) Equipment
Detailed Design Review: P18001
Presentation transcript:

Power Partitioning for Multiprocessors Kai Ma, Amiri-Kamalabad Mojtaba

Background Problems:  Overall power: electricity bill (date center), battery life (embedded system)  On-chip component power: Irregular power characteristics of each components lead to conservative cooling and packaging design. Power capping or control  Minimizing performance degradation on the condition of capping or controlling power (MICRO06 Isci, HPCA07 Dybdahl, PACT08 Meng)

Motivation Different applications have different power consumption patterns. Power consumption in one application could change dramatically. If adopting pessimistic design, we prepare for peak power dissipation. Most of the time, extra cooling capability will be wasted.

Proposal & Novelties We propose a scheme to control each on-chip component’s power and partition the overall power budget to each component to optimize performance.  Universalize dramatically different components’ characteristics to power consumption and performance contribution trade off, solve the problem in systematic framework (previous: trial-and-error)  Provide upstream and downstream flexibility (previous: case by case)

Design Challenges 1. Power Control for each component 1) Regulate each component’s power behavior 2) Guarantee the whole chip power Challenges: 1) How to guarantee system stability? 2. Power budget partitioning Challenges: 1) How to find the optimal partitioning? 2) How to adapt to different workloads or different execution phase for one workload?

Possible System Design

Initial Result Feedback control can achieve power control goal. Different power allocations lead to different performance.