Yuxiang Li

Using Artificial Neural Network for Predicting Thread Partitioning in Speculative Multithreading

Speculative multithreading (SpMT) is a thread-level automatic parallelization technique to accelerate sequential programs on multi-core, and it partitions programs into multiple threads to be speculatively executed in the presence of ambiguous data and control dependences while the correctness of the programs is guaranteed by hardware support. Thread granularity, number of parallel threads as well as partition postions are crucial to the performance improvement in SpMT, for they determine the amount of resources (CPU, memory, cache, or waiting cycles, etc), and affect the efficiency of every PE (Processing Element). In conventional way, these three parameters are determined by heuristic rules. Although it is simple to partition threads with them, they are a type of one-size-fits-all strategy and can not guarantee to get the optimal solution of thread partitioning. This paper proposes an Artificial Neural Network (ANN) based approach to learn and determine the thread partition strategy. Using the ANN-based thread partition approach, an unseen irregular program can obtain a stable, much higher speedup than the Heuristic Rules (HR) based approach. On Prophet, which is a generic SpMT processor to evaluate the performance of multithreaded programs, the novel thread partitioning policy is evaluated and reaches an average speedup of 1.80 on 4-core processor. Experiments show that our proposed approach can obtain a significant increase in speedup and Olden benchmarks deliver a better performance improvement of 2.36% than the traditional heuristic rules based approach. The results indicate that our approach finds the best partitioning scheme for each program and is more stable across programs.

A Hybrid Samples Generation Approach in Speculative Multithreading

Speculative Multithreading (SpMT) is a thread level automatic parallelization technique to accelerate sequential programs. Machine learning has been successfully introduced into SpMT to improve its performance. An appropriate sample set is important for machine learning-based (ML-based) thread partition. Conventionally, heuristic rules-based (HR-based) sample generation approach can not generate adaptive samples, which conform to ML-based thread partition platform and characteristics of source applications. A hybrid sample generation approach is proposed. With this method, we firstly automatically generate samples which are mips codes consisting spawning points (SP) and control quasi-independent points (CQIP) by heuristic rules, then manually adjust the positions of SP and CQIP and rebuild pre-computation slice (p-slice) to obtain better performance for every application. During the implementation of this approach, three measures: bias weighting, reservation of optimal solutions, summary of greedy rules are carried out. In this way, we enhance the adjustment frequency for the subroutine with high called number, achieving a stable improvement of speedups. On Prophet, which is a generic SpMT processor to evaluate the performance of multithreaded programs, Olden benchmarks are used to realize the approach. Experiments show that the approach can obtain better sample sets, which deliver a better performance improvement of about 10% on a 32 core than conventional heuristic-based approach. Experiments also prove that this approach is effective to get sample sets for ML-based thread partition methods.

Optimization of thread partitioning parameters in speculative multithreading based on artificial immune algorithm

Thread partition plays an important role in speculative multithreading (SpMT) for automatic parallelization of irregular programs. Using unified values of partition parameters to partition different applications leads to the fact that every application cannot own its optimal partition scheme. In this paper, five parameters affecting thread partition are extracted from heuristic rules. They are the dependence threshold (DT), lower limit of thread size (TSL), upper limit of thread size (TSU), lower limit of spawning distance (SDL), and upper limit of spawning distance (SDU). Their ranges are determined in accordance with heuristic rules, and their step-sizes are set empirically. Under the condition of setting speedup as an objective function, all combinations of five threshold values form the solution space, and our aim is to search for the best combination to obtain the best thread granularity, thread dependence, and spawning distance, so that every application has its best partition scheme. The issue can be attributed to a single objective optimization problem. We use the artificial immune algorithm (AIA) to search for the optimal solution. On Prophet, which is a generic SpMT processor to evaluate the performance of multithreaded programs, Olden benchmarks are used to implement the process. Experiments show that we can obtain the optimal parameter values for every benchmark, and Olden benchmarks partitioned with the optimized parameter values deliver a performance improvement of 3.00% on a 4-core platform compared with a machine learning based approach, and 8.92% compared with a heuristics-based approach.

Similarity Assessment of Program Samples Based on Theory of Fuzzy

Using machine learning has proven effective at choosing the right set of optimizations for a particular program. For machine learning techniques to be most effective, compiler writers have to develop expressive means of characterizing the program being optimized. The start-of-art techniques for characterizing programs include using a fixed-length feature vector of either source code features extracted during compile time or performance counters collected when running the program. According to the program features, similarity values are calculated to assess the similar degree. In this paper, we introduce a novel way of assessing the similarity of two program samples using Theory of Fuzzy. We firstly calculate the Euclidean Distance of two different program samples as the input, and then assess the overall similarity degree as well as respective similarity degree, using corresponding Fuzzy Function. The graph-based characterization technique is cited, which shows great advantages than state-of-the-art characterization techniques found in the literature. By using the sequences predicted to be the best by the graph-based model, we compare the known sequences with unknown ones, and calculate the similarity between them, then obtain the similarity degree. Using these degrees, we are directed to make better determination to cluster or classify.

A hybrid sample generation approach in speculative multithreading

Speculative multithreading (SpMT) is a thread-level automatic parallelization technique to accelerate sequential programs. Machine learning has been successfully brought into SpMT to improve its performance. An appropriate sample set, which plays the role of knowledge provider, is important for machine learning-based (ML-based) thread partition. Conventionally, heuristic rules-based (HR-based) sample generation approach cannot generate adaptive samples. A hybrid sample generation approach can break this bottleneck. With this method, we firstly automatically generate samples, which are MIPS codes consisting of spawning points (SPs) and control quasi-independent points (CQIPs) by heuristic rules; secondly manually adjust the positions of SPs and CQIPs and rebuild pre-computation slice to obtain better performance for every sample; and then build model to ensure that the probability of adjusting to the optimal partition positions is increasing. During the implementation of this approach, three measures: bias weighting, preservation of optimal solutions, summary of greedy rules, are taken. In this way, we enhance the adjustment frequency for subroutines with high called time and preserve the optimal partition positions, so to achieve a stable speedup improvement. On Prophet, which is a generic SpMT processor to evaluate the performance of multithreaded programs, SPEC2000 and Olden benchmarks are used as input. Experiments show that our approach can obtain better sample sets, which deliver a better performance improvement of about 86.9% on a 16 core than the samples generated by HR-based approach. Experiment results also prove that this approach is effective to generate sample sets for ML-based thread partition.

A speculative parallel decompression algorithm on Apache Spark

Data decompression is one of the most important techniques in data processing and has been widely used in multimedia information transmission and processing. However, the existing decompression algorithms on multicore platforms are time-consuming and do not support large data well. In order to expand parallelism and enhance decompression efficiency on large-scale datasets, based on the software thread-level speculation technique, this paper raises a speculative parallel decompression algorithm on Apache Spark. By analyzing the data structure of the compressed data, the algorithm firstly hires a function to divide compressed data into blocks which can be decompressed independently and then spawns a number of threads to speculatively decompress data blocks in parallel. At last, the speculative results are merged to form the final outcome. Comparing with the conventional parallel approach on multicore platform, the proposed algorithm is very efficiency and obtains a high parallelism degree by making the best of the resources of the cluster. Experiments show that the proposed approach could achieve 2.6

GbA: A graph‐based thread partition approach in speculative multithreading

\times

Similar Samples Cleaning in Speculative Multithreading

× speedup when comparing with the traditional approach in average. In addition, with the growing number of working nodes, the execution time cost decreases gradually, and the speedup scales linearly. The results indicate that the decompression efficiency can be significantly enhanced by adopting this speculative parallel algorithm.