Matches in SemOpenAlex for { <https://semopenalex.org/work/W2567051523> ?p ?o ?g. }
- W2567051523 endingPage "13" @default.
- W2567051523 startingPage "1" @default.
- W2567051523 abstract "Cache capacity and memory bandwidth play critical roles in application performance, particularly for data-intensive applications from domains that include machine learning, numerical analysis, and data mining. Many of these applications are also tolerant to imprecise inputs and have loose constraints on the quality of output, making them ideal candidates for approximate computing. This paper introduces a novel approximate computing technique that decouples the format of data in the memory hierarchy from the format of data in the compute subsystem to significantly reduce the cost of storing and moving bits throughout the memory hierarchy and improve application performance. This asymmetric compute-memory extension to conventional architectures, ACME, adds two new instruction classes to the ISA — load-concise and store-concise — along with three small functional units to the micro-architecture to support these instructions. ACME does not affect exact execution of applications and comes into play only when concise memory operations are used. Through detailed experimentation we find that ACME is very effective at trading result accuracy for improved application performance. Our results show that ACME achieves a 1.3x speedup (up to 1.8x) while maintaining 99% accuracy, or a 1.1x speedup while maintaining 99.999% accuracy. Moreover, our approach incurs negligible area and power overheads, adding just 0.005% area and 0.1% power to a conventional modern architecture." @default.
- W2567051523 created "2017-01-06" @default.
- W2567051523 creator A5002075773 @default.
- W2567051523 creator A5009277466 @default.
- W2567051523 creator A5041306152 @default.
- W2567051523 creator A5044722987 @default.
- W2567051523 creator A5053236545 @default.
- W2567051523 creator A5058184922 @default.
- W2567051523 creator A5065128438 @default.
- W2567051523 creator A5073136757 @default.
- W2567051523 creator A5080959197 @default.
- W2567051523 date "2016-10-15" @default.
- W2567051523 modified "2023-09-22" @default.
- W2567051523 title "Concise loads and stores: the case for an asymmetric compute-memory architecture for approximation" @default.
- W2567051523 cites W1567646530 @default.
- W2567051523 cites W1967519200 @default.
- W2567051523 cites W1976919080 @default.
- W2567051523 cites W2020517863 @default.
- W2567051523 cites W2026005150 @default.
- W2567051523 cites W2033982707 @default.
- W2567051523 cites W2048588974 @default.
- W2567051523 cites W2048791451 @default.
- W2567051523 cites W2068965783 @default.
- W2567051523 cites W2069738303 @default.
- W2567051523 cites W2080592089 @default.
- W2567051523 cites W2080794684 @default.
- W2567051523 cites W2081368694 @default.
- W2567051523 cites W2097243222 @default.
- W2567051523 cites W2099517310 @default.
- W2567051523 cites W2101756288 @default.
- W2567051523 cites W2102449048 @default.
- W2567051523 cites W2105544671 @default.
- W2567051523 cites W2107399212 @default.
- W2567051523 cites W2109884165 @default.
- W2567051523 cites W2111444234 @default.
- W2567051523 cites W2114703523 @default.
- W2567051523 cites W2119299853 @default.
- W2567051523 cites W2124608923 @default.
- W2567051523 cites W2128120785 @default.
- W2567051523 cites W2135393827 @default.
- W2567051523 cites W2141955346 @default.
- W2567051523 cites W2143283746 @default.
- W2567051523 cites W2147657366 @default.
- W2567051523 cites W2147657598 @default.
- W2567051523 cites W2162752393 @default.
- W2567051523 cites W2170382128 @default.
- W2567051523 cites W2187230075 @default.
- W2567051523 cites W2233797083 @default.
- W2567051523 cites W2237415205 @default.
- W2567051523 cites W2411972054 @default.
- W2567051523 cites W2464177207 @default.
- W2567051523 cites W2950769435 @default.
- W2567051523 cites W2963374099 @default.
- W2567051523 cites W2963674932 @default.
- W2567051523 doi "https://doi.org/10.5555/3195638.3195688" @default.
- W2567051523 hasPublicationYear "2016" @default.
- W2567051523 type Work @default.
- W2567051523 sameAs 2567051523 @default.
- W2567051523 citedByCount "10" @default.
- W2567051523 countsByYear W25670515232017 @default.
- W2567051523 countsByYear W25670515232018 @default.
- W2567051523 countsByYear W25670515232019 @default.
- W2567051523 countsByYear W25670515232021 @default.
- W2567051523 countsByYear W25670515232022 @default.
- W2567051523 crossrefType "proceedings-article" @default.
- W2567051523 hasAuthorship W2567051523A5002075773 @default.
- W2567051523 hasAuthorship W2567051523A5009277466 @default.
- W2567051523 hasAuthorship W2567051523A5041306152 @default.
- W2567051523 hasAuthorship W2567051523A5044722987 @default.
- W2567051523 hasAuthorship W2567051523A5053236545 @default.
- W2567051523 hasAuthorship W2567051523A5058184922 @default.
- W2567051523 hasAuthorship W2567051523A5065128438 @default.
- W2567051523 hasAuthorship W2567051523A5073136757 @default.
- W2567051523 hasAuthorship W2567051523A5080959197 @default.
- W2567051523 hasConcept C115537543 @default.
- W2567051523 hasConcept C118524514 @default.
- W2567051523 hasConcept C123657996 @default.
- W2567051523 hasConcept C142362112 @default.
- W2567051523 hasConcept C153349607 @default.
- W2567051523 hasConcept C162324750 @default.
- W2567051523 hasConcept C173608175 @default.
- W2567051523 hasConcept C176649486 @default.
- W2567051523 hasConcept C188045654 @default.
- W2567051523 hasConcept C21547014 @default.
- W2567051523 hasConcept C2776257435 @default.
- W2567051523 hasConcept C2778100165 @default.
- W2567051523 hasConcept C2778915421 @default.
- W2567051523 hasConcept C2779602883 @default.
- W2567051523 hasConcept C2781357197 @default.
- W2567051523 hasConcept C31170391 @default.
- W2567051523 hasConcept C31258907 @default.
- W2567051523 hasConcept C34447519 @default.
- W2567051523 hasConcept C41008148 @default.
- W2567051523 hasConcept C68339613 @default.
- W2567051523 hasConcept C9390403 @default.
- W2567051523 hasConcept C98986596 @default.
- W2567051523 hasConceptScore W2567051523C115537543 @default.
- W2567051523 hasConceptScore W2567051523C118524514 @default.