CSE 591: Energy-Efficient Computing Lecture 19 SPEED: memory

Slides:



Advertisements
Similar presentations
55:035 Computer Architecture and Organization Lecture 7 155:035 Computer Architecture and Organization.
Advertisements

International Symposium on Low Power Electronics and Design Energy-Efficient Non-Minimal Path On-chip Interconnection Network for Heterogeneous Systems.
CSE 691: Energy-Efficient Computing Lecture 20 review Anshul Gandhi 1307, CS building
1 MemScale: Active Low-Power Modes for Main Memory Qingyuan Deng, David Meisner*, Luiz Ramos, Thomas F. Wenisch*, and Ricardo Bianchini Rutgers University.
COMPARISON OF ADAPTIVE VOLTAGE/FREQUENCY SCALING AND ASYNCHRONOUS PROCESSOR ARCHITECTURES FOR NEURAL SPIKE SORTING J. Leverett A. Pratt R. Hochman May.
CSE 691: Energy-Efficient Computing Lecture 4 SCALING: stateless vs. stateful Anshul Gandhi 1307, CS building
CSE 531: Performance Analysis of Systems Lecture 1: Intro and Logistics Anshul Gandhi 1307, CS building
Memory Power Management via Dynamic Voltage/Frequency Scaling Howard David (Intel) Eugene Gorbatov (Intel) Ulf R. Hanebutte (Intel) Chris Fallin (CMU)
CS 7810 Lecture 20 Initial Observations of the Simultaneous Multithreading Pentium 4 Processor N. Tuck and D.M. Tullsen Proceedings of PACT-12 September.
CS 7810 Lecture 12 Power-Aware Microarchitecture: Design and Modeling Challenges for Next-Generation Microprocessors D. Brooks et al. IEEE Micro, Nov/Dec.
Memory Redundancy Elimination to Improve Application Energy Efficiency Keith Cooper and Li Xu Rice University October 2003.
Computer ArchitectureFall 2007 © October 31, CS-447– Computer Architecture M,W 10-11:20am Lecture 17 Review.
Embedded DRAM for a Reconfigurable Array S.Perissakis, Y.Joo 1, J.Ahn 1, A.DeHon, J.Wawrzynek University of California, Berkeley 1 LG Semicon Co., Ltd.
CSE 691: Energy-Efficient Computing Lecture 3 SLEEP: full-system Anshul Gandhi 1307, CS building
1 Enabling Large Scale Network Simulation with 100 Million Nodes using Grid Infrastructure Hiroyuki Ohsaki Graduate School of Information Sci. & Tech.
CSE 691: Energy-Efficient Computing Lecture 6 SHARING: distributed vs. local Anshul Gandhi 1307, CS building
CSE 691: Energy-Efficient Computing Lecture 7 SMARTS: custom-made systems Anshul Gandhi 1307, CS building
Critical Power Slope Understanding the Runtime Effects of Frequency Scaling Akihiko Miyoshi, Charles Lefurgy, Eric Van Hensbergen Ram Rajamony Raj Rajkumar.
E X C E E D I N G E X P E C T A T I O N S Time Mgt Linux System Administration Dr. Hoganson Kennesaw State University OS Time Management Time management.
Oracle Tuning Considerations. Agenda Why Tune ? Why Tune ? Ways to Improve Performance Ways to Improve Performance Hardware Hardware Software Software.
Performance David Monismith Jan. 16, 2015 Based on notes from Dr. Bill Siever and from the Patterson and Hennessy Text.
Lessons from HLT benchmarking (For the new Farm) Rainer Schwemmer, LHCb Computing Workshop 2014.
A Row Buffer Locality-Aware Caching Policy for Hybrid Memories HanBin Yoon Justin Meza Rachata Ausavarungnirun Rachael Harding Onur Mutlu.
CSE 691: Energy-Efficient Computing Lecture 1: Intro and Logistics Anshul Gandhi 1307, CS building
VGreen: A System for Energy Efficient Manager in Virtualized Environments G. Dhiman, G Marchetti, T Rosing ISLPED 2009.
Supporting Systolic and Memory Communication in iWarp CS258 Paper Summary Computer Science Jaein Jeong.
Dynamic Voltage Frequency Scaling for Multi-tasking Systems Using Online Learning Gaurav DhimanTajana Simunic Rosing Department of Computer Science and.
CS-7081 Irreflexive – (13 - 8). CS-7082 Example – (13 - 9)
A Measurement Based Memory Performance Evaluation of Streaming Media Servers Garba Isa Yau and Abdul Waheed Department of Computer Engineering King Fahd.
1 Lecture 2: Memory Energy Topics: energy breakdowns, handling overfetch, LPDRAM, row buffer management, channel energy, refresh energy.
CSE 591: Energy-Efficient Computing Lecture 3 SPEED: processor Anshul Gandhi 347, CS building
CSE 591: Energy-Efficient Computing Lecture 1: Intro and Logistics Anshul Gandhi 347, New CS building
CSE 591: Energy-Efficient Computing Lecture 4 SLEEP: full-system Anshul Gandhi 347, CS building
CSE 591: Energy-Efficient Computing Lecture 8 SOURCE: renewables Anshul Gandhi 347, CS building
Equalizer: Dynamically Tuning GPU Resources for Efficient Execution Ankit Sethia* Scott Mahlke University of Michigan.
© 2008 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice ProLiant G5 to G6 Processor Positioning.
M AESTRO : Orchestrating Predictive Resource Management in Future Multicore Systems Sangyeun Cho, Socrates Demetriades Computer Science Department University.
CSE 591: Energy-Efficient Computing Lecture 13 SLEEP: sensors
Computer Organization Exam Review CS345 David Monismith.
PPEP: online Performance, power, and energy prediction framework
Network Connected Multiprocessors
Anshul Gandhi 347, CS building
Anshul Gandhi 347, CS building
Exploiting Inter-Warp Heterogeneity to Improve GPGPU Performance
CSE 591: Energy-Efficient Computing Lecture 17 SCALING: survey
What to do when an exam question has data…
CSE 591: Energy-Efficient Computing Lecture 20 SPEED: disks
Decoupled Access-Execute Pioneering Compilation for Energy Efficiency
CSE 591: Energy-Efficient Computing Lecture 21 review
CSE 591: Energy-Efficient Computing Lecture 15 SCALING: storage
CSE 544, Spring 2018 Probability and Statistics for Data Science Lecture 1: Intro and Logistics Anshul Gandhi 347, New CS building
BitWarp Energy Efficient Analytic Data Processing on Next Generation General Purpose GPUs Jason Power || Yinan Li || Mark D. Hill || Jignesh M. Patel.
CSE 591: Energy-Efficient Computing Lecture 10 SLEEP: network
Understanding and Exploiting Amazon EC2 Spot Instances
Optimal Power Allocation in Server Farms
25.1 The Electromagnetic Spectrum
Lecture 2: Performance Today’s topics: Technology wrap-up
CSE 591: Energy-Efficient Computing Lecture 12 SLEEP: memory
CSE 591: Energy-Efficient Computing Lecture 14 SCALING: setup time
CSE 591: Energy-Efficient Computing Lecture 9 SLEEP: processor
CS 140 Lecture Notes: Technology and Operating Systems
Computer Architecture
CS 140 Lecture Notes: Technology and Operating Systems
CSE 531: Performance Analysis of Systems Lecture 4: DTMC
CSE 591: Energy-Efficient Computing Lecture 18 SPEED: power
If a DRAM has 512 rows and its refresh time is 9ms, what should be the frequency of row refresh operation on the average?
Advanced Computer Networks
Performance Metrics Neil Tang 9/10/2008
A Novel Cache-Utilization Based Dynamic Voltage Frequency Scaling (DVFS) Mechanism for Reliability Enhancements *Yen-Hao Chen, *Yi-Lun Tang, **Yi-Yu Liu,
Power-Aware DVFS on PowerPC 405LP: Front Bus Scaling
Presentation transcript:

CSE 591: Energy-Efficient Computing Lecture 19 SPEED: memory Anshul Gandhi 347, CS building anshul@cs.stonybrook.edu

memscale paper

Memory subsystem

Memory power breakdown MEM MIX CPU

Remarks on memory DVFS DVFS lowers power, but effect on energy is not obvious as response time might increase Some memory access stages get delayed, not necessarily end-to-end latencies

Performance monitoring top sar collectl perf stat

Energy management policy | 5 millisecs | 300 microsecs

Controller

Controller Oscillations?

Evaluation

Comparisons

Sensitivity analysis Similar to? DreamWeaver

memory_dvfs paper

Memory power %

Active energy

Background power

VF scaling

Latency vs. B/W Similar to what queueing theory suggests.

B/W vs. savings Compared to 1333 MHz.

Algo in action Measure bandwidth in previous epoch and find frequency for next epoch based on observed latency vs bandwidth graph.