Memory power optimisation on low-bit multi-access cross memory address mapping schema
by Zongwei Zhu; Xi Li; Chao Wang; Xuehai Zhou
International Journal of Embedded Systems (IJES), Vol. 6, No. 2/3, 2014

Abstract: Since memory accounts for an increasing fraction of computer energy, manufacturers have developed memory devices with different power/work modes and lots of creditable power mode control algorithms have been proposed. By analysing these policies on different memory address mapping schemas (schema is used to translate a physical address to a memory cell), we find they are less effective on low-bit multi-access cross memory (LMCM) schema which is a fine-grained interleaving schema to improve parallelism to meet high-bandwidth. For gaining more power efficiency, we introduce a comprehensive solution named as MSPA. It adopts a memory address segmentation module (MASM) to split memory into many regions configured as different schemas. With the help of an OS power-aware memory allocator (PAMA), MSPA can allocate one applications memory from its preferred region to balance power and performance. Finally, by integrating some outstanding works, experimental results demonstrate that it can further improve power efficiency from 3% to 17%.

Online publication date: Thu, 31-Jul-2014

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Embedded Systems (IJES):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com