Analysis of static and dynamic energy consumption in NUCA caches

Abstract

NUCA caches are large L2 on-chip cache memories characterized by multi-bank partitioning and designed to hide wire delay effects. They exhibit high hit rates while keeping access latency low. Proposed designs for such caches are Static NUCA, in which data are statically allocated to the cache banks, and Dynamic NUCA, in which data may reside in different banks, and a migration mechanism is introduced to better tolerate wire delay effects. The two architectures permit to achieve different performances by acting on architectural parameters and data management policies, at the cost of different balances between static and dynamic power consumption and energy dissipation. In this work, we propose preliminary results of the characterization of such balances, by presenting an evaluation of performance and energy consumption of conventional UCAs, and Static and Dynamic NUCA caches. All the considered caches architectures are equal sized and they are supposed to be used in an aggressive high frequency system running some applications from the SPEC CPU2000 and the NAS Parallel Benchmarks suites. The experimental results obtained indicate that, although the migration of data contributes to increase the dynamic energy consumption in Dynamic NUCA caches, the higher IPC achieved permits to save static energy, which dominates the power/energy balance in all the considered architectures. As a consequence, such results would designate NUCA caches as the most performing and energy saving architectures. Besides, according to the obtained results, future power improvements for NUCA caches should concentrate on static energy, while, for the dynamic energy, the on-chip network is the most critical element. Migration of data is acceptable, since it has a positive impact on performance, and the increased dynamic energy is overwhelmed by the static energy savings resulting from the shorter execution time. In order to give a general validity to such statements, we need to explore more design space points for each architecture (by varying the running clock rate and other design parameters) and to evaluate them considering a larger set of benchmarks.

An Adaptive Block Pinning Cache for Reducing Network Traffic in Multi-core Architectures

“…We have observed that different applications have different cache block size requirements and it effect cache miss rate which is directly correlated with the performance and the size of the data transfer using interconnect network with current and future wire-limited technologies [2] effect bandwidth which can be directly correlated with dynamic energy [8]. For example, the influences of block granularity on miss rate and bandwidth for a 64K L1 cache and a 1M L2 cache keeping the number of ways constant to accommodate Group A of applications, when the block size is changed from 64B to 32B the miss rate increases by 2x times for the group B.…”

A power-efficient migration mechanism for D-NUCA caches

“…Chishti et al [3] have proposed an optimization scheme, called NuRAPID, aiming at increasing the energy efficiency and the performance of NUCA caches in a single core configuration. An analysis of the various components of static and dynamic power consumption has been performed in [4]. In [5] a lightweight hardware technique which reduces static power consumption for D-NUCA architectures is proposed.…”

LRU-PEA: A smart replacement policy for non-uniform cache architectures on chip multiprocessors

“…We deal with a similar energy model to that adopted by Bardine et al [9]. Therefore, we also consider the static and dynamic energy dissipated by the NUCA cache and the additional energy required to access the off-chip memory.…”