Methodology for operation shuffling and L0 cluster generation for low energy heterogeneous VLIW processors

Abstract

Clustering L0 buffers is effective for energy reduction in the instruction memory hierarchy of embedded VLIW processors. However, the efficiency of the clustering depends on the schedule of the target application. Especially in heterogeneous or data clustered VLIW processors, determining energy efficient scheduling is more constraining.This article proposes a realistic technique supported by a tool flow to explore operation shuffling for improving generation of L0 clusters. The tool flow explores assignment of operations for each cycle and generates various schedules. This approach makes it possible to reduce energy consumption for various processor architectures. However, the computational complexity is large because of the huge exploration space. Therefore, some heuristics are also developed, which reduce the size of the exploration space while the solution quality remains reasonable. Furthermore, we also propose a technique to support VLIW processors with multiple data clusters, which is essential to apply the methodology to real world processors.The experimental results indicate potential gains of up to 27.6% in energy in L0 buffers, through operation shuffling for heterogeneous processor architectures as well as a homogeneous This work is supported by JSPS Grant-in-Aid for JSPS Fellows #17.9887. Based on Kobayashi, Y., Jayapala, M., Raghavan, P., Catthoor, F., and Imai, M. Operation shuffling for low energy L0 cluster generation on heterogeneous VLIW processors.

Operation shuffling over cycle boundaries for low energy L0 clustering

“…Slot capability of 10 slot heterogeneous VLIW processor. Slot 0 1 2 3 4 5 6 7 8 9 alu * * * * * * shift * * * * * * mult/div * * fp alu * * * * * * fp mult/div * * load/store * * branch * * * * * * posed shuffling employs heuristics for both of vertical shuffling and horizontal shuffling proposed in [6]. Table 3 shows experimental results.…”

“…Here we can say assignment (3), (5), (6), and (7) are the (2) and (5), cycle 0, 3, and 6 generate apparently the same variety of slot assignment. However, cycle 1 in (3) and cycle 4 in (5) also generate the same variety because both cycles has one operation, 'e'.…”

“…Therefore, operation shuffling technique [6] is proposed, where operations are tried to move to different slot then a lot of schedules are generated. By optimizing L0 cluster configuration for each schedule and estimating energy for it, the most energy efficient schedule are selected among the schedules.…”

“…Even in the horizontal shuffling, the exploration space is hard to treat unless some heuristics are introduced [6]. Then the vertical shuffling would generate much more schedules since each operation can move not only to another slot but also to another cycle.…”

“…it is complicated to determine an energy efficient schedule. For improving the energy efficiency of L0 clusters, operation shuffling is proposed [6], which explores assignment of operations for each cycle, generates various schedules, and evaluates them to find an energy efficient schedule. Some heuristics are also proposed to reduce the size of the exploration space.…”