Large-scale CMPs with hundreds of cores require a directory-based protocol to maintain cache coherence. However, previously proposed coherence directories are hard to scale beyond tens of cores, requiring either excessive area or energy, complex hierarchical protocols, or inexact representations of sharer sets that increase coherence traffic and degrade performance. We present SCD, a scalable coherence directory that relies on efficient highly-associative caches (such as zcaches) to implement a single-level directory that scales to thousands of cores, tracks sharer sets exactly, and incurs negligible directory-induced invalidations. SCD scales because, unlike conventional directories, it uses a variable number of directory tags to represent sharer sets: lines with one or few sharers use a single tag, while widely shared lines use additional tags, so tags remain small as the system scales up. We show that, thanks to the efficient highly-associative array it relies on, SCD can be fully characterized using analytical models, and can be sized to guarantee a negligible number of evictions independently of the workload. We evaluate SCD using simulations of a 1024-core CMP. For the same level of coverage, we find that SCD is 13× more area-efficient than full-map sparse directories, and 2× more area-efficient and faster than hierarchical directories, while requiring a simpler protocol. Furthermore, we show that SCD's analytical models are accurate in practice.

/pdf/scd-a-scalable-coherence-directory-with-flexible-sharer-set-419bt1438h.pdf

SCD: A scalable coherence directory with flexible sharer set encoding

Over the last years Transactional Memory (TM) gained growing popularity as a simpler, attractive alternative to classic lock-based synchronization schemes. Recently, the TM landscape has been profoundly changed by the integration of Hardware TM (HTM) in Intel commodity processors, raising a number of questions on the future of TM. We seek answers to these questions by conducting the largest study on TM to date, comparing different locking techniques, hardware and software TMs, as well as different combinations of these mechanisms, from the dual perspective of performance and power consumption. Our study sheds a mix of light and shadows on currently available commodity HTM: on one hand, we identify workloads in which HTM clearly outperforms any alternative synchronization mechanism; on the other hand, we show that current HTM implementations suffer of restrictions that narrow the scope in which these can be more effective than state of the art software solutions. Thanks to the results of our study, we identify a number of compelling research problems in the areas of TM design, compilers and self-tuning.

/pdf/virtues-and-limitations-of-commodity-hardware-transactional-31bq3jgejc.pdf

Virtues and limitations of commodity hardware transactional memory

Embedded-TM: Energy and complexity-effective hardware transactional memory for embedded multicore systems

Two overriding concerns in the development of embedded MPSoCs are ease of programming and hardware complexity. In this paper we present SoC-TM, an integrated HW/SW solution for transactional programming on embedded MP-SoCs. Our proposal leverages a Hardware Transactional Memory (HTM) design, based on a dedicated HW module for conflict management, whose functionality is exposed to the software through compiler directives, implemented as an extension to the popular OpenMP programming model. To further improve ease of programming, our framework supports speculative parallelism, thanks to the ability of enforcing a given commit order in hardware. Our experimental results confirm that SoC-TM is a viable and cost-effective solution for embedded MPSoCs, in terms of energy, performance and productivity.

SoC-TM: integrated HW/SW support for transactional memory programming on embedded MPSoCs

Transactional Memory is currently being advocated as a promising alternative to lock-based synchronization because it simplifies multithreaded programming. In this way, future many-core CMP architectures may need to provide hardware support for transactional memory. On the other hand, power dissipation constitutes a first class consideration in multicore processor design. 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 the Log TM-SE Eager-Eager system and a version of the Scalable TCC Lazy-Lazy system that enables parallel commits. To the best of our knowledge, this is the first characterization in terms of energy consumption of hardware transactional memory systems. To do that, we extended the 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 60% higher on average than in the Lazy-Lazy case, whereas performance differences between the two systems are 42% on average. Finally, we found that although on average Lazy-Lazy beats Eager-Eager there are considerable deviations in performance depending on the particular characteristics of each application.

/pdf/characterizing-energy-consumption-in-hardware-transactional-4i40wa8t6p.pdf

Characterizing Energy Consumption in Hardware Transactional Memory Systems

We propose a new design for an energy-efficient hardware transactional memory (HTM) system for power-aware embedded devices. Prior hardware transactional memory designs proposed a small, fully-associative transactional cache at the same level as the L1 cache. We propose an alternative design that unifies the transactional and L1 caches, and provides a small victim cache to reduce effects of capacity and conflict evictions. We evaluate our new HTM scheme on a variety of benchmarks, both in terms of energy and performance. We show that the victim cache scheme can provide up to a 4X improvement in energy-delay product, compared to a traditional HTM scheme that uses a separate transactional cache.

Samantha Wood

Papers

Embedded-TM: Energy and complexity-effective hardware transactional memory for embedded multicore systems

Energy and throughput efficient transactional memory for embedded multicore systems