TY - GEN
T1 - Energy efficient last level caches via last read/write prediction
AU - Alves, Marco A.Z.
AU - Villavieja, Carlos
AU - Diener, Matthias
AU - Navaux, Philippe O.A.
PY - 2013
Y1 - 2013
N2 - The size of the Last Level Caches (LLC) in multicore architectures is increasing, and so is their power consumption. However, most of this power is wasted on unused or invalid cache lines. For dirty cache lines, the LLC waits until the line is evicted to be written back to memory. Hence, dirty lines compete for the memory bandwidth with read requests (prefetch and demand), increasing pressure on the memory controller. This paper proposes a Dead Line and Early Write-Back Predictor (DEWP) to improve the energy efficiency of the LLC. DEWP early evicts dead cache lines with an average accuracy of 94%, and only 2% false positives. DEWP also allows scheduling of dirty lines for early eviction, allowing earlier write-backs. Using DEWP over a set of single and multi-threaded benchmarks, we obtain an average of 61% static energy savings, while maintaining the performance, for both inclusive and non-inclusive LLCs.
AB - The size of the Last Level Caches (LLC) in multicore architectures is increasing, and so is their power consumption. However, most of this power is wasted on unused or invalid cache lines. For dirty cache lines, the LLC waits until the line is evicted to be written back to memory. Hence, dirty lines compete for the memory bandwidth with read requests (prefetch and demand), increasing pressure on the memory controller. This paper proposes a Dead Line and Early Write-Back Predictor (DEWP) to improve the energy efficiency of the LLC. DEWP early evicts dead cache lines with an average accuracy of 94%, and only 2% false positives. DEWP also allows scheduling of dirty lines for early eviction, allowing earlier write-backs. Using DEWP over a set of single and multi-threaded benchmarks, we obtain an average of 61% static energy savings, while maintaining the performance, for both inclusive and non-inclusive LLCs.
UR - http://www.scopus.com/inward/record.url?scp=84893582258&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84893582258&partnerID=8YFLogxK
U2 - 10.1109/SBAC-PAD.2013.12
DO - 10.1109/SBAC-PAD.2013.12
M3 - Conference contribution
AN - SCOPUS:84893582258
SN - 9781479929276
T3 - Proceedings - Symposium on Computer Architecture and High Performance Computing
SP - 73
EP - 80
BT - Proceedings - 2013 25th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2013
PB - IEEE Computer Society
T2 - 2013 25th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2013
Y2 - 23 October 2013 through 26 October 2013
ER -