Power/performance trade-offs in real-time SDRAM command scheduling

S.L.M. Goossens, K. Chandrasekar, K.B. Akesson, K.G.W. Goossens

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

7 Citaties (Scopus)
91 Downloads (Pure)

Uittreksel

Real-time safety-critical systems should provide hard bounds on an applications’ performance. SDRAM controllers used in this domain should therefore have a bounded worst-case bandwidth, response time, and power consumption. Existing works on real-time SDRAM controllers only consider a narrow range of memory devices, and do not evaluate how their schedulers’ performance varies across memory generations, nor how the scheduling algorithm influences power usage. The extent to which the number of banks used in parallel to serve a request impacts performance is also unexplored, and hence there are gaps in the tool set of a memory subsystem designer, in terms of both performance analysis, and configuration options. This article introduces a generalized close-page memory command scheduling algorithm that uses a variable number of banks in parallel to serve a request. To reduce the schedule length for DDR4 memories, we exploit bank grouping through a pairwise bank-group interleaving scheme. The algorithm is evaluated using an ILP formulation, and provides schedules of optimal length for most of the considered LPDDR, DDR2, DDR3, LPDDR2, LPDDR3 and DDR4 devices. We derive the worst-case bandwidth, power and execution time for the same set of devices, and discuss the observed trade-offs and trends in the scheduler-configuration design space based on these metrics, across memory generations.
Originele taal-2Engels
Pagina's (van-tot)1882-1895
TijdschriftIEEE Transactions on Computers
Volume56
Nummer van het tijdschrift6
DOI's
StatusGepubliceerd - 1 jun 2016

Vingerafdruk

Trade-offs
Scheduling
Real-time
Data storage equipment
Scheduling algorithms
Scheduler
Scheduling Algorithm
Schedule
Bandwidth
Controller
Inductive logic programming (ILP)
Safety-critical Systems
Controllers
Configuration
Interleaving
Grouping
Execution Time
Response Time
Power Consumption
Performance Analysis

Citeer dit

Goossens, S.L.M. ; Chandrasekar, K. ; Akesson, K.B. ; Goossens, K.G.W. / Power/performance trade-offs in real-time SDRAM command scheduling. In: IEEE Transactions on Computers. 2016 ; Vol. 56, Nr. 6. blz. 1882-1895.
@article{b4a6a191b15e4d1684a128be48a5a29d,
title = "Power/performance trade-offs in real-time SDRAM command scheduling",
abstract = "Real-time safety-critical systems should provide hard bounds on an applications’ performance. SDRAM controllers used in this domain should therefore have a bounded worst-case bandwidth, response time, and power consumption. Existing works on real-time SDRAM controllers only consider a narrow range of memory devices, and do not evaluate how their schedulers’ performance varies across memory generations, nor how the scheduling algorithm influences power usage. The extent to which the number of banks used in parallel to serve a request impacts performance is also unexplored, and hence there are gaps in the tool set of a memory subsystem designer, in terms of both performance analysis, and configuration options. This article introduces a generalized close-page memory command scheduling algorithm that uses a variable number of banks in parallel to serve a request. To reduce the schedule length for DDR4 memories, we exploit bank grouping through a pairwise bank-group interleaving scheme. The algorithm is evaluated using an ILP formulation, and provides schedules of optimal length for most of the considered LPDDR, DDR2, DDR3, LPDDR2, LPDDR3 and DDR4 devices. We derive the worst-case bandwidth, power and execution time for the same set of devices, and discuss the observed trade-offs and trends in the scheduler-configuration design space based on these metrics, across memory generations.",
author = "S.L.M. Goossens and K. Chandrasekar and K.B. Akesson and K.G.W. Goossens",
year = "2016",
month = "6",
day = "1",
doi = "10.1109/TC.2015.2458859",
language = "English",
volume = "56",
pages = "1882--1895",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "6",

}

Power/performance trade-offs in real-time SDRAM command scheduling. / Goossens, S.L.M.; Chandrasekar, K.; Akesson, K.B.; Goossens, K.G.W.

In: IEEE Transactions on Computers, Vol. 56, Nr. 6, 01.06.2016, blz. 1882-1895.

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

TY - JOUR

T1 - Power/performance trade-offs in real-time SDRAM command scheduling

AU - Goossens, S.L.M.

AU - Chandrasekar, K.

AU - Akesson, K.B.

AU - Goossens, K.G.W.

PY - 2016/6/1

Y1 - 2016/6/1

N2 - Real-time safety-critical systems should provide hard bounds on an applications’ performance. SDRAM controllers used in this domain should therefore have a bounded worst-case bandwidth, response time, and power consumption. Existing works on real-time SDRAM controllers only consider a narrow range of memory devices, and do not evaluate how their schedulers’ performance varies across memory generations, nor how the scheduling algorithm influences power usage. The extent to which the number of banks used in parallel to serve a request impacts performance is also unexplored, and hence there are gaps in the tool set of a memory subsystem designer, in terms of both performance analysis, and configuration options. This article introduces a generalized close-page memory command scheduling algorithm that uses a variable number of banks in parallel to serve a request. To reduce the schedule length for DDR4 memories, we exploit bank grouping through a pairwise bank-group interleaving scheme. The algorithm is evaluated using an ILP formulation, and provides schedules of optimal length for most of the considered LPDDR, DDR2, DDR3, LPDDR2, LPDDR3 and DDR4 devices. We derive the worst-case bandwidth, power and execution time for the same set of devices, and discuss the observed trade-offs and trends in the scheduler-configuration design space based on these metrics, across memory generations.

AB - Real-time safety-critical systems should provide hard bounds on an applications’ performance. SDRAM controllers used in this domain should therefore have a bounded worst-case bandwidth, response time, and power consumption. Existing works on real-time SDRAM controllers only consider a narrow range of memory devices, and do not evaluate how their schedulers’ performance varies across memory generations, nor how the scheduling algorithm influences power usage. The extent to which the number of banks used in parallel to serve a request impacts performance is also unexplored, and hence there are gaps in the tool set of a memory subsystem designer, in terms of both performance analysis, and configuration options. This article introduces a generalized close-page memory command scheduling algorithm that uses a variable number of banks in parallel to serve a request. To reduce the schedule length for DDR4 memories, we exploit bank grouping through a pairwise bank-group interleaving scheme. The algorithm is evaluated using an ILP formulation, and provides schedules of optimal length for most of the considered LPDDR, DDR2, DDR3, LPDDR2, LPDDR3 and DDR4 devices. We derive the worst-case bandwidth, power and execution time for the same set of devices, and discuss the observed trade-offs and trends in the scheduler-configuration design space based on these metrics, across memory generations.

U2 - 10.1109/TC.2015.2458859

DO - 10.1109/TC.2015.2458859

M3 - Article

VL - 56

SP - 1882

EP - 1895

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 6

ER -