Leakage power in cache memories represents a sizable fraction of total power consumption, and many techniques have been proposed to reduce it. As a matter of fact, during a fixed period of time, only a small subset of cache lines is used. Previous techniques put unused lines, for example, to drowsy in order to save power. Our idea is to adaptively select the most used cache lines. In the case of instruction cache, we found that this can automatically achieved by coupling a tiny cache acting as a filter cache (IL0 cache) with a drowsy-cache. Our experiments, with complete MiBench suite for ARM based processor, show a 25% improvement in leakage saving versus drowsy.
Giorgi, R., Bennati, P. (2008). Filtering drowsy instruction cache to achieve better efficiency. In Proceedings of the 23th ACM Symposium on Applied Computing (pp.1554-1555). ACM [10.1145/1363686.1364050].
Filtering drowsy instruction cache to achieve better efficiency
GIORGI, ROBERTO;
2008-01-01
Abstract
Leakage power in cache memories represents a sizable fraction of total power consumption, and many techniques have been proposed to reduce it. As a matter of fact, during a fixed period of time, only a small subset of cache lines is used. Previous techniques put unused lines, for example, to drowsy in order to save power. Our idea is to adaptively select the most used cache lines. In the case of instruction cache, we found that this can automatically achieved by coupling a tiny cache acting as a filter cache (IL0 cache) with a drowsy-cache. Our experiments, with complete MiBench suite for ARM based processor, show a 25% improvement in leakage saving versus drowsy.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11365/46866
Attenzione
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo