Abstract
With the increased complexity of platforms coupled with data centers’ servers sprawl, power consumption is reaching unsustainable limits. Memory is an important target for platform-level energy efficiency, where most power management techniques use multiple power state DRAM devices to transition them to low-power states when they are “sufficiently” idle. However, fully-interleaved memory in high-performance servers presents a research challenge to the memory power management problem. Due to data striping across all memory modules, memory accesses are distributed in a manner that considerably reduces the idleness of memory modules to warrant transitions to low-power states. In this paper we introduce a novel technique for dynamic memory interleaving that is adaptive to incoming workload in a manner that reduces memory energy consumption while maintaining the performance at an acceptable level. We use optimization theory to formulate and solve the power-performance management problem. We use dynamic cache line migration techniques to increase the idleness of memory modules by consolidating the application’s working-set on a minimal set of ranks. Our technique yields energy saving of about 48.8 % (26.7 kJ) compared to traditional techniques measured at 4.5%. It delivers the maximum performance-per-watt during all phases of the application execution with a maximum performance-per-watt improvement of 88.48%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Lebeck, R., et al.: Power aware page allocation. In: Proc. ASPLOS-9 (2000)
Rambus, RDRAM (1999), http://www.rambus.com
DDR2 FBDIMM Technical Product Specifications
Fan, X., et al.: Memory controller policies for dram power management. In: Proc. ISLPED (2001)
Zhou, P., et al.: Dynamic tracking of page miss ratio curve for memory management. In: Proc. ASPLOS-11 (2004)
SPECjbb (2005), http://www.spec.org/jbb2005/docs/WhitePaper.html
Bovet, D., Cesati, M.: Understanding the Linux Kernel, O’Reilly (2002)
Delaluz, V., et al.: Hardware and Software Techniques for Controlling DRAM Power Modes. IEEE Trans. Computers 50(11), 1154–1173 (2001)
Delaluz, V., et al.: Scheduler-Based DRAM Energy Management. In: Proc. DAC-39 (2002)
Huang, H., et al.: Design and Implementation of Power-Aware Virtual Memory. In: Proc. USENIX Technical Conference, pp. 57–70 (2003)
Delaluz, et al.: Automatic Data Migration for Reducing Energy Consumption in Multi-Bank Memory Systems. In: Proc. 39th Design Automation Conf., pp. 213–218. ACM Press, New York (2002)
Wang, D., et al.: DRAMsim: A memory-system simulator. SIGARCH Computer Architecture News 33(4), 100–107 (2005)
Li, et al.: Performance-directed energy management for main memory and disks. In: Proc. ASPLOS-11, Boston, MA, USA (2004)
Diniz, et al.: Limiting the power consumption of main memory. In: Proc. ISCA-34, San Diego, California, USA (2007)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Khargharia, B., Hariri, S., Yousif, M.S. (2007). Self-optimization of Performance-per-Watt for Interleaved Memory Systems. In: Aluru, S., Parashar, M., Badrinath, R., Prasanna, V.K. (eds) High Performance Computing – HiPC 2007. HiPC 2007. Lecture Notes in Computer Science, vol 4873. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-77220-0_35
Download citation
DOI: https://doi.org/10.1007/978-3-540-77220-0_35
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-77219-4
Online ISBN: 978-3-540-77220-0
eBook Packages: Computer ScienceComputer Science (R0)