Enabling scalability-sensitive speculative parallelization for FSM computations

Abstract

Finite state machines (FSMs) are the backbone of many applications, but are difficult to parallelize due to their inherent dependencies. Speculative FSM parallelization has shown promise on multicore machines with up to eight cores. However, as hardware parallelism grows (e.g., Xeon Phi has up to 288 logical cores), a fundamental question raises: How does the speculative FSM parallelization scale as the number of cores increases? Without answering this question, existing methods for speculative FSM parallelization simply choose to use all available cores, which might not only waste computing resources, but also result in suboptimal performance. In this work, we conduct a systematic scalability analysis for speculative FSM parallelization. Unlike many other parallelizations which can be modeled by the classic Amdahl's law or its simple extensions, speculative FSM parallelization scales unconventionally due to the non-deterministic nature of speculation and the cost variations of misspeculation. To address these challenges, this work introduces a spectrum of scalability models that are customized to the properties of specific FSMs and the underlying architecture. The models, for the first time, precisely capture the scalability of speculative parallelization for different FSM computations, and clearly show the existence of a "sweet spot" in terms of the number of cores employed by the speculative FSM parallelization to achieve the optimal performance. To make the scalability models practical, we develop S3, a scalability-sensitive speculation framework for FSM parallelization. For any given FSM, S3 can automatically characterize its properties and analyze its scalability, hence guide speculative parallelization towards the optimal performance and more efficient use of computing resources. Evaluations on different FSMs and architectures confirm the accuracy of the proposed models and show that S3 achieves significant speedup (up to 5X) and energy savings (up to 77%). CCS CONCEPTS • Computing methodologies → Parallel algorithms; • Computer systems organization → Parallel architectures;

Challenging Sequential Bitstream Processing via Principled Bitwise Speculation

“…If the verification fails, the corresponding input partition has to be reprocessed with the correct starting state. Despite some optimization [53] for stopping the reprocessing earlier, the reprocessing cost, in general, can significantly compromise the speculation benefits [37].…”

Scalable Processing of Contemporary Semi-Structured Data on Commodity Parallel Processors - A Compilation-based Approach

“…An implementation of this method [50] has been optimized for SIMD instructions. Speculative parallelization of FSMs has also been exploited [56,57,64,65] by breaking the transition dependences with state predictions. Though providing useful insights, the above work cannot be directly applied to the parallelization of semi-structured data processing, which essentially requires the use of stack-based automata.…”

Enabling prefix sum parallelism pattern for recurrences with principled function reconstruction

“…Many previous research works [11,16,17,25,26] have demonstrated the effectiveness of state-convergence for finite state machine algorithms (FSM). According to their studies, most FSMs, even those with many states, often converge to 16 or less active states for any input.…”

“…Zhao et al [25,26] parallelized a set of finite state machines (FSMs) with principled speculation. Qiu et al [17] observed that speculative FSM parallelization is not scalable with an increasing number of cores. To address the limitation, they presented a series of scalability analysis models and developed an automatic speculative FSM parallelization framework named S3.…”