Toward Dark Silicon in Servers
IEEE MicroPublished 1 July 2011Open access
Nikos Hardavellas, Michael Ferdman, Babak Falsafi, Anastasia Ailamaki
Citations311
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
Server chips will not scale beyond a few tens to low hundreds of cores, and an increasing fraction of the chip in future technologies will be dark silicon that the authors cannot afford to power.
Abstract
Server chips will not scale beyond a few tens to low hundreds of cores, and an increasing fraction of the chip in future technologies will be dark silicon that we cannot afford to power. Specialized multicore processors, however, can leverage the underutilized die area to overcome the initial power barrier, delivering significantly higher performance for the same bandwidth and power envelopes.
Keywords
Computer ScienceEngineering
Dark silicon and the end of multicore scaling
1,511 Citations2011Hadi Esmaeilzadeh, Emily Blem +3 more
The study shows that regardless of chip organization and topology, multicore scaling is power limited to a degree not widely appreciated by the computing community.
IEEE Journal of Solid-State CircuitsA dynamic voltage scaled microprocessor system
903 Citations2000Tom Burd, Trevor Pering +2 more
An adaptive, non-uniform cache structure for wire-delay dominated on-chip caches
736 Citations2002Changkyu Kim, Doug Burger +1 more
This paper proposes physical designs for these Non-Uniform Cache Architectures (NUCAs) and extends these physical designs with logical policies that allow important data to migrate toward the processor within the same level of the cache.
Dynamic Warp Formation and Scheduling for Efficient GPU Control Flow
727 Citations2007Wilson Wai Lun Fung, Ivan Sham +2 more
This work implements two major extensions to the CACTI cache modeling tool that focus on interconnect design for a large cache, and adopts state-of-the-art design space exploration strategies for non-uniform cache access (NUCA).
ACM SIGARCH Computer Architecture News3D-Stacked Memory Architectures for Multi-core Processors
543 Citations2008Gabriel H. Loh
This work explores more aggressive 3D DRAM organizations that make better use of the additional die-to-die bandwidth provided by 3D stacking, as well as the additional transistor count, to achieve a 1.75x speedup over previously proposed 3D-DRAM approaches on memory-intensive multi-programmed workloads on a quad-core processor.
Optimizing NUCA Organizations and Wiring Alternatives for Large Caches with CACTI 6.0
482 Citations2007Xiaoyao Liang, Ramón Canal +1 more
A range of cache refresh and placement schemes that are sensitive to retention time are proposed, and it is shown that most of the retention time variations can be masked by the microarchitecture when using these schemes.
3D-Stacked Memory Architectures for Multi-core Processors
433 Citations2008Gabriel H. Loh
Understanding sources of inefficiency in general-purpose chips
393 Citations2010Rehan Hameed, Wajahat Qadeer +7 more
The sources of performance and energy overheads in general-purpose processing systems are explored by quantifying the overheads of a 720p HD H.264 encoder running on a general-purpose CMP system and methods to eliminate these overheads are explored by transforming the CPU into a specialized system for H.264 encoding.
IEEE MicroSimFlex: Statistical Sampling of Computer System Simulation
334 Citations2006Thomas F. Wenisch, Roland E. Wunderlich +4 more
Statistical sampling makes simulation-based studies feasible by providing ten-thousand-fold reductions in simulation runtime and enabling thousand-way simulation parallelism.
Single-Chip Heterogeneous Computing: Does the Future Include Custom Logic, FPGAs, and GPGPUs?
245 Citations2010Eric S. Chung, Peter Milder +2 more
The model considers the less-obvious relationships between conventional processors and a diverse set of U-cores to understand the relative merits between different approaches in the face of technology constraints, and supports speculation of future designs from scaling trends predicted by the ITRS road map.
A dynamic voltage scaled microprocessor system
198 Citations2000Tom Burd, Trevor Pering +2 more
If clock frequency and supply voltage are dynamically varied in response to computational load demands, then energy consumed per process can be reduced for the low computational periods, while retaining peak performance when required, called dynamic voltage scaling (DVS).
Database Servers on Chip Multiprocessors: Limitations and Opportunities
157 Citations2007Nikos Hardavellas, Ippokratis Pandis +3 more
This paper characterize the performance of a commercial database server running on emerging chip multiprocessor technologies and finds that the major bottleneck of current software is data cache stalls, with L2 hit stalls rising from oblivion to become the dominant execution time component in some cases.
IEEE MicroThe GreenDroid Mobile Application Processor: An Architecture for Silicon's Dark Future
147 Citations2011Nathan Goulding-Hotta, Jack Sampson +10 more
The Greendroid mobile application processor demonstrates an approach that uses dark silicon to execute general-purpose smart phone applications with less energy than today's most energy efficient designs.
Maximizing CMP throughput with mediocre cores
145 Citations2005John D. Davis, James Laudon +1 more
For the best overall CMT performance, smaller cores with lower performance, so called "mediocre" cores, maximize the total number of CMT cores and outperform CMTs built from larger, higher performance cores.
IEEE Journal of Solid-State CircuitsA Power-Efficient High-Throughput 32-Thread SPARC Processor
136 Citations2007Ana Sonia Leon, Kenway W. Tam +3 more
The first generation of Niagara SPARC processors implements a power-efficient multi-threading architecture to achieve high throughput with minimum hardware complexity.
Energy/power breakdown of pipelined nanometer caches (90nm/65nm/45nm/32nm)
55 Citations2006Samuel Rodríguez, Bruce Jacob
It is shown that subthreshold leakage is the main cause of static power dissipation, and that gate leakage is, surprisingly, not a significant contributor to total cache power, even for 32nm caches.
WiDGET
51 Citations2010Yasuko Watanabe, John D. Davis +1 more
WiDGET's decoupled design provides flexibility to alter resource allocation for a particular power-performance target while turning off unallocated resources, and enables dynamic customization of different combinations of small and/or powerful cores on a single chip, consuming power in proportion to the delivered performance.
A Power-Efficient High-Throughput 32-Thread SPARC Processor
40 Citations2006Ana Sonia Leon, Jinuk Luke Shin +6 more
This first generation of "Niagara" SPARC processors implements a power-efficient Chip Multi-Threading (CMT) architecture which maximizes overall throughput performance for commercial workloads and reduces hardware complexity and power.
Performance Trend in Three-Dimensional Integrated Circuits
13 Citations2006Hao Hua, Christopher Mineo +4 more
This work develops an automated design flow with 2D CAD tools to design 3DICs with the MIT Lincoln Lab 0.18mum three-tier fully depleted silicon on insulator (FDSOI) process, showing that the performance of3DIC shows up to two-generation advantage over its 2D counterpart with only three tiers.
Infoscience (Ecole Polytechnique Fédérale de Lausanne)Power Scaling: the Ultimate Obstacle to 1K-Core Chips
12 Citations2010Nikos Hardavellas, Michael Ferdman +2 more
This work explores the design space of physically-constrained multicore chips across technologies and shows that, even with conservative estimates, chips will not scale beyond a few tens of cores due to physical power and off-chip bandwidth constraints, potentially leaving the die real-estate underutilized in future technology generations.
Chip multiprocessors for server workloads
10 Citations2009Babak Falsafi, Anastasia Ailamaki +1 more
Reactive NUCA (R-NUCA), a distributed cache design which reacts to the class of each access to place blocks close to the requesting cores is proposed, and it is found that heterogeneous multicores hold great promise in improving designs even further.