On the design of energy‐efficient hardware transactional memory systems

Abstract

SUMMARYTransactional memory is currently being advocated as a promising alternative to lock‐based synchronization because it simplifies multithreaded programming. In this way, future many‐core chip multiprocessor architectures may need to provide hardware support for transactional memory. On the other hand, energy consumption constitutes nowadays a first class consideration in multicore processor designs. In this work, we characterize the performance and energy consumption of two well‐known hardware transactional memory systems that employ opposite policies for data versioning and conflict management. More specifically, we compare a LogTM‐SE eager‐eager system and a version of the Scalable Transactional Coherence and Consistency lazy‐lazy system that enable parallel commits. To do so, we extended the Multifacet GEMS simulator to estimate the energy consumed in the on‐chip caches according to CACTI and used the interconnection network energy model given by Orion 2. Results show that the energy consumption of the eager‐eager system is 38% higher in average than in the lazy‐lazy case, whereas performance differences between the two systems are 26% in average. We found that even though lazy‐lazy beats eager‐eager on average, there are considerable deviations in performance depending on the particular characteristics of each application and the settings of both systems. Finally, from this characterization, we observe that a significant part of the energy consumed in some applications in eager‐eager is spent on the back‐off delay phase and explore more energy‐efficient hardware back‐off mechanisms. For lazy‐lazy systems, the way in which memory lines are assigned to the L2 cache banks affects the number of parallel commits in some applications, and we study an alternative fine‐grained assignment. Copyright © 2012 John Wiley & Sons, Ltd.

Virtues and limitations of commodity hardware transactional memory

“…In both [21,20] the authors assess the behaviour of different HTM implementations via simulation (the latter focusing on embedded systems). The approach was also taken by [2], where the power consumption of one STM was studied via simulation.…”

An Energy Computation in Distributed Computing Environment through Bellman-Ford Algorithm

“…Ballga et al [4] have described analysis and technology for energy consumption of data throughput in network's like DSL, HFC networks, passive optical networks, point to point optical systems, W-CDMA, WiMAX. Gaona et al [5] have given the design of energy-efficient in hardware transactional memory systems. Chiaraviglio et al [6] have given the model based energy consumption mechanism such as device architecture and load for networks.…”

Using Frequency Scaling on Virtualized Memory in Cloud Datacenters

“…In addition, memory energy management is challenging in the context of servers with modern DRAM technologies. According to the recent studies [6,7], main memory in massive server platforms accounts for up to 40% of server energy which is comparable to or slightly higher than the energy consumption contribution of CPUs. In cloud datacenters, the fraction attributable to memory accesses may be even higher, because resource virtualization technology will introduce extra energy…”