Creator |
|
|
|
Language |
|
Publisher |
|
|
Date |
|
Source Title |
|
Source Title |
|
Vol |
|
Issue |
|
First Page |
|
Last Page |
|
Publication Type |
|
Access Rights |
|
Rights |
|
Related DOI |
|
Related DOI |
|
|
|
Related URI |
|
|
|
Related URI |
|
Related HDL |
|
Relation |
|
|
|
Abstract |
これまでに多くの低リーク・キャッシュが提案された.しかしながら,これらの手法は待機状態ラインのデータを破棄するため,ミス回数が増加し必然的に性能が低下する.そこで本稿では低リーク・キャッシュにおける性能低下抑制方式として,常活性ライン方式を提案する.具体的には,性能低下の原因となる待機状態ライン・アクセスの局所性を考慮し,アクセスが集中するラインは常活性ラインにする.これまでに提案されたCache... decay方式では,15.1%程度の性能低下をもたらす事で92.7%のリーク削減率を達成した.これに対し,本稿で提案する方式を適用すると,同程度のリーク削減率90.6%を維持しつつ,性能低下を5.0%に抑制することができた. A number of techniques to reduce cache leakage energy have so far been proposed. However, in these techniques, flushing the data of a turning off line causes a new cache miss. And, the increase miss degrade processor performance. We have analyzed the detail of cache-access behavior, and have found that there is a locality of accesses to the turning-off lines. Based on this observation, we propose a cache management technique to alleviate the negative effect of low-leakage caches. In our approach, cache lines having high degree of increase-miss locality are forced to stay in the high-speed but high-leakage mode. In our evaluation, the proposed scheme worsens the performance by only 5.0% with the same degree of energy reduction of the Cache decay approach.show more
|