Clustering L0 buffers is effective for energy reduction in the instruction memory caches of embedded VLIW processors. However, the efficiency of the clustering depends on the schedule of the target application. For improving the energy efficiency of L0 clusters, an operation shuffling is proposed, which explores assignment of operations for each cycle, generates various schedules, and evaluates them to find an energy efficient schedule. This approach can find energy efficient schedules, however, it takes a long time to obtain the final result. In this paper, we propose a new method to directly generate an energy efficient schedule without iterations of operation shuffling. In the proposed method, a compiler schedules operations using the result of the single operation shuffling as a constraint. We propose some optimization algorithms to generate an energy efficient schedule for a given L0 cluster organization. The proposed method can drastically reduce the computational effort since it performs the operation shuffling only once. The experimental results show that comparable energy reduction is achieved by using the proposed method while the computational effort can be reduced significantly over the conventional operation shuffling.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yuki KOBAYASHI, Murali JAYAPALA, Praveen RAGHAVAN, Francky CATTHOOR, Masaharu IMAI, "Efficient Method to Generate an Energy Efficient Schedule Using Operation Shuffling" in IEICE TRANSACTIONS on Fundamentals,
vol. E91-A, no. 2, pp. 604-612, February 2008, doi: 10.1093/ietfec/e91-a.2.604.
Abstract: Clustering L0 buffers is effective for energy reduction in the instruction memory caches of embedded VLIW processors. However, the efficiency of the clustering depends on the schedule of the target application. For improving the energy efficiency of L0 clusters, an operation shuffling is proposed, which explores assignment of operations for each cycle, generates various schedules, and evaluates them to find an energy efficient schedule. This approach can find energy efficient schedules, however, it takes a long time to obtain the final result. In this paper, we propose a new method to directly generate an energy efficient schedule without iterations of operation shuffling. In the proposed method, a compiler schedules operations using the result of the single operation shuffling as a constraint. We propose some optimization algorithms to generate an energy efficient schedule for a given L0 cluster organization. The proposed method can drastically reduce the computational effort since it performs the operation shuffling only once. The experimental results show that comparable energy reduction is achieved by using the proposed method while the computational effort can be reduced significantly over the conventional operation shuffling.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1093/ietfec/e91-a.2.604/_p
Copy
@ARTICLE{e91-a_2_604,
author={Yuki KOBAYASHI, Murali JAYAPALA, Praveen RAGHAVAN, Francky CATTHOOR, Masaharu IMAI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Efficient Method to Generate an Energy Efficient Schedule Using Operation Shuffling},
year={2008},
volume={E91-A},
number={2},
pages={604-612},
abstract={Clustering L0 buffers is effective for energy reduction in the instruction memory caches of embedded VLIW processors. However, the efficiency of the clustering depends on the schedule of the target application. For improving the energy efficiency of L0 clusters, an operation shuffling is proposed, which explores assignment of operations for each cycle, generates various schedules, and evaluates them to find an energy efficient schedule. This approach can find energy efficient schedules, however, it takes a long time to obtain the final result. In this paper, we propose a new method to directly generate an energy efficient schedule without iterations of operation shuffling. In the proposed method, a compiler schedules operations using the result of the single operation shuffling as a constraint. We propose some optimization algorithms to generate an energy efficient schedule for a given L0 cluster organization. The proposed method can drastically reduce the computational effort since it performs the operation shuffling only once. The experimental results show that comparable energy reduction is achieved by using the proposed method while the computational effort can be reduced significantly over the conventional operation shuffling.},
keywords={},
doi={10.1093/ietfec/e91-a.2.604},
ISSN={1745-1337},
month={February},}
Copy
TY - JOUR
TI - Efficient Method to Generate an Energy Efficient Schedule Using Operation Shuffling
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 604
EP - 612
AU - Yuki KOBAYASHI
AU - Murali JAYAPALA
AU - Praveen RAGHAVAN
AU - Francky CATTHOOR
AU - Masaharu IMAI
PY - 2008
DO - 10.1093/ietfec/e91-a.2.604
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E91-A
IS - 2
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - February 2008
AB - Clustering L0 buffers is effective for energy reduction in the instruction memory caches of embedded VLIW processors. However, the efficiency of the clustering depends on the schedule of the target application. For improving the energy efficiency of L0 clusters, an operation shuffling is proposed, which explores assignment of operations for each cycle, generates various schedules, and evaluates them to find an energy efficient schedule. This approach can find energy efficient schedules, however, it takes a long time to obtain the final result. In this paper, we propose a new method to directly generate an energy efficient schedule without iterations of operation shuffling. In the proposed method, a compiler schedules operations using the result of the single operation shuffling as a constraint. We propose some optimization algorithms to generate an energy efficient schedule for a given L0 cluster organization. The proposed method can drastically reduce the computational effort since it performs the operation shuffling only once. The experimental results show that comparable energy reduction is achieved by using the proposed method while the computational effort can be reduced significantly over the conventional operation shuffling.
ER -