ESKIMO: Energy savings using Semantic Knowledge of Inconsequential Memory Occupancy for DRAM subsystem
Dynamic Random Access Memory (DRAM) is used as the bulk of the main memory in most computing systems and its energy and power consumption has become a first-class design consideration for modern systems. We propose ESKIMO, a scheme where when the program or operating systems memory manager allocates...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Text |
Language: | English |
Published: |
2009
|
Subjects: | |
Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.396.9683 http://people.cs.pitt.edu/~cho/cs2410/papers/isen-micro09.pdf |
id |
ftciteseerx:oai:CiteSeerX.psu:10.1.1.396.9683 |
---|---|
record_format |
openpolar |
spelling |
ftciteseerx:oai:CiteSeerX.psu:10.1.1.396.9683 2023-05-15T16:07:22+02:00 ESKIMO: Energy savings using Semantic Knowledge of Inconsequential Memory Occupancy for DRAM subsystem Ciji Isen Lizy John The Pennsylvania State University CiteSeerX Archives 2009 application/pdf http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.396.9683 http://people.cs.pitt.edu/~cho/cs2410/papers/isen-micro09.pdf en eng http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.396.9683 http://people.cs.pitt.edu/~cho/cs2410/papers/isen-micro09.pdf Metadata may be used without restrictions as long as the oai identifier remains attached to it. http://people.cs.pitt.edu/~cho/cs2410/papers/isen-micro09.pdf and energy savings. General Terms Management Performance Design Experimentation. Keywords Memory power and energy cross-boundary or cross-layer architecture optimizations allocated text 2009 ftciteseerx 2016-01-08T02:31:57Z Dynamic Random Access Memory (DRAM) is used as the bulk of the main memory in most computing systems and its energy and power consumption has become a first-class design consideration for modern systems. We propose ESKIMO, a scheme where when the program or operating systems memory manager allocates or frees up a memory region, this information is used by the architecture to optimize the working of the DRAM system, particularly to save energy and power. In this work we attempt to have the architecture work hand in hand with information about allocated and freed space provided by the program. We discuss multiple ways to use this information to reduce the energy and power consumption of the memory and present results of this optimization. We evaluate the energy and power benefits of our technique using a publicly available, hardware-validated, DRAM simulator, DRAMsim [1]. Our current studies show very promising results with energy savings on average of 39%. Text eskimo* Unknown Freed ENVELOPE(164.333,164.333,-71.483,-71.483) |
institution |
Open Polar |
collection |
Unknown |
op_collection_id |
ftciteseerx |
language |
English |
topic |
and energy savings. General Terms Management Performance Design Experimentation. Keywords Memory power and energy cross-boundary or cross-layer architecture optimizations allocated |
spellingShingle |
and energy savings. General Terms Management Performance Design Experimentation. Keywords Memory power and energy cross-boundary or cross-layer architecture optimizations allocated Ciji Isen Lizy John ESKIMO: Energy savings using Semantic Knowledge of Inconsequential Memory Occupancy for DRAM subsystem |
topic_facet |
and energy savings. General Terms Management Performance Design Experimentation. Keywords Memory power and energy cross-boundary or cross-layer architecture optimizations allocated |
description |
Dynamic Random Access Memory (DRAM) is used as the bulk of the main memory in most computing systems and its energy and power consumption has become a first-class design consideration for modern systems. We propose ESKIMO, a scheme where when the program or operating systems memory manager allocates or frees up a memory region, this information is used by the architecture to optimize the working of the DRAM system, particularly to save energy and power. In this work we attempt to have the architecture work hand in hand with information about allocated and freed space provided by the program. We discuss multiple ways to use this information to reduce the energy and power consumption of the memory and present results of this optimization. We evaluate the energy and power benefits of our technique using a publicly available, hardware-validated, DRAM simulator, DRAMsim [1]. Our current studies show very promising results with energy savings on average of 39%. |
author2 |
The Pennsylvania State University CiteSeerX Archives |
format |
Text |
author |
Ciji Isen Lizy John |
author_facet |
Ciji Isen Lizy John |
author_sort |
Ciji Isen |
title |
ESKIMO: Energy savings using Semantic Knowledge of Inconsequential Memory Occupancy for DRAM subsystem |
title_short |
ESKIMO: Energy savings using Semantic Knowledge of Inconsequential Memory Occupancy for DRAM subsystem |
title_full |
ESKIMO: Energy savings using Semantic Knowledge of Inconsequential Memory Occupancy for DRAM subsystem |
title_fullStr |
ESKIMO: Energy savings using Semantic Knowledge of Inconsequential Memory Occupancy for DRAM subsystem |
title_full_unstemmed |
ESKIMO: Energy savings using Semantic Knowledge of Inconsequential Memory Occupancy for DRAM subsystem |
title_sort |
eskimo: energy savings using semantic knowledge of inconsequential memory occupancy for dram subsystem |
publishDate |
2009 |
url |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.396.9683 http://people.cs.pitt.edu/~cho/cs2410/papers/isen-micro09.pdf |
long_lat |
ENVELOPE(164.333,164.333,-71.483,-71.483) |
geographic |
Freed |
geographic_facet |
Freed |
genre |
eskimo* |
genre_facet |
eskimo* |
op_source |
http://people.cs.pitt.edu/~cho/cs2410/papers/isen-micro09.pdf |
op_relation |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.396.9683 http://people.cs.pitt.edu/~cho/cs2410/papers/isen-micro09.pdf |
op_rights |
Metadata may be used without restrictions as long as the oai identifier remains attached to it. |
_version_ |
1766403454691966976 |