Run-time power-down strategies for real-time SDRAM memory controllers

Powering down SDRAMs at run-time reduces memory energy consumption significantly, but often at the cost of performance. If employed speculatively with real-time memory controllers, power-down mechanisms could impact both the guaranteed bandwidth and the memory latency bounds. This calls for power-down strategies that can hide or bound the performance loss, making run-time memory power-down feasible for real-time applications. In this paper, we propose two such strategies that reduce memory energy consumption and yet guarantee realtime memory performance. One provides significant energy savings without impacting the guaranteed bandwidth and latency bounds. The other provides higher energy savings with marginally increased latency bounds, while still preserving the guaranteed bandwidth provided by real-time memory controllers. We also present an algorithm to select the most energy-efficient power-down mode at run-time. We experimentally evaluate the two strategies at run-time by executing four media applications concurrently on a real-time MPSoC platform and show memory energy savings of 42.1% and 51.3% for the two strategies, respectively.

[1]  Kees Goossens,et al.  Memory Controllers for Real-Time Embedded Systems , 2012 .

[2]  Francisco J. Cazorla,et al.  An Analyzable Memory Controller for Hard Real-Time CMPs , 2009, IEEE Embedded Systems Letters.

[3]  Artur Burchard,et al.  A real-time streaming memory controller , 2005, Design, Automation and Test in Europe.

[4]  Mahmut T. Kandemir,et al.  Hardware and Software Techniques for Controlling DRAM Power Modes , 2001, IEEE Trans. Computers.

[5]  Anujan Varma,et al.  Latency-rate servers: a general model for analysis of traffic scheduling algorithms , 1996, Proceedings of IEEE INFOCOM '96. Conference on Computer Communications.

[6]  Todd M. Austin,et al.  The SimpleScalar tool set, version 2.0 , 1997, CARN.

[7]  Kees G. W. Goossens,et al.  Architectures and modeling of predictable memory controllers for improved system integration , 2011, 2011 Design, Automation & Test in Europe.

[8]  Edward A. Lee,et al.  PRET DRAM controller: Bank privatization for predictability and temporal isolation , 2011, 2011 Proceedings of the Ninth IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS).

[9]  Kees G. W. Goossens,et al.  Improved Power Modeling of DDR SDRAMs , 2011, 2011 14th Euromicro Conference on Digital System Design.

[10]  Christof Pitter,et al.  Time-predictable memory arbitration for a Java chip-multiprocessor , 2008, JTRES '08.

[11]  K. B. Akesson,et al.  Memory-map selection for firm real-time memory controllers , 2012 .

[12]  Calvin Lin,et al.  A comprehensive approach to DRAM power management , 2008, 2008 IEEE 14th International Symposium on High Performance Computer Architecture.

[13]  Bruce Jacob,et al.  Memory Systems: Cache, DRAM, Disk , 2007 .

[14]  Kees G. W. Goossens,et al.  CoMPSoC: A template for composable and predictable multi-processor system on chips , 2009, TODE.

[15]  Stephen A. Edwards,et al.  A disruptive computer design idea: Architectures with repeatable timing , 2009, 2009 IEEE International Conference on Computer Design.

[16]  Kees Goossens,et al.  Memory Controllers for Real-Time Embedded Systems: Predictable and Composable Real-Time Systems , 2011 .