To better support data-intensive workflows which are typically built out of various independently developed executables, this paper proposes extensions to parallel database systems called User-Defined eXecutables (UDX) and collective queries. UDX facilitates the description of workflows by enabling seamless integrations of external executables into SQL statements without any efforts to write programs confirming to strict specifications of databases. A collective query is an SQL query whose results are distributed to multiple clients and then processed by them in parallel, using arbitrary UDX. It provides efficient parallelization of executables through the data transfer optimization algorithms that distribute query results to multiple clients, taking both communication cost and computational loads into account. We implement this concept in a system called ParaLite, a parallel database system based on a popular lightweight database SQLite. Our experiments show that ParaLite has several times higher performance over Hive for typical SQL tasks and has 10x speedup compared to a commercial DBMS for executables. In addition, this paper studies a real-world text processing workflow and builds it on top of ParaLite, Hadoop, Hive and general files. Our experiences indicate that ParaLite outperforms other systems in both productivity and performance for the workflow.

With the consideration of real-time stream processing technology, it's important to develop high availability mechanism to guarantee stream-based application not interfered by faults caused by potential anomalies. In this paper, we present a novel online prediction technique for predicting some anomalies which may occur in the near future. Concretely, we first present a value prediction which combines the Hidden Markov Model and the Mixture of Expert Model to predict the values of feature metrics in the near future. Then we employ the Support Vector Machine to do anomaly identification, which is a procedure to identify the kind of anomaly that we are about to alarm. The purpose of our approach is to achieve a tradeoff between fault penalty and resource cost. The experiment results show that our approach is of high accuracy for common anomaly prediction and low runtime overhead.

The original single-shot multibox detector (SSD) algorithm has good detection accuracy and speed for regular object recognition. However, the SSD is not suitable for detecting small objects for two reasons: 1) the relationships among different feature layers with various scales are not considered, 2) the predicted results are solely determined by several independent feature layers. To enhance its detection capability for small objects, this study proposes an improved SSD-based algorithm called proportional channels' fusion SSD (PCF-SSD). Three enhancements are provided by this novel PCF-SSD algorithm. First, a fusion feature pyramid model is proposed by concatenating channels of certain key feature layers in a given proportion for object detection. Second, the default box sizes are adjusted properly for small object detection. Third, an improved loss function is suggested to train the above-proposed fusion model, which can further improve object detection performance. A series of experiments are conducted on the public database Pascal VOC to validate the PCF-SSD. On comparing with the original SSD algorithm, our algorithm improves the mean average precision and detection accuracy for small objects by 3.3% and 3.9%, respectively, with a detection speed of 40FPS. Furthermore, the proposed PCF-SSD can achieve a better balance of detection accuracy and efficiency than the original SSD algorithm, as demonstrated by a series of experimental results.

Data exchange, in which two blocks of data are swapped between cores in distributed memory systems, necessitates additional memory buffer in a multiprocessor system-on-chip. In this paper, we propose a novel bidirectional inter-core communication mechanism called coherent direct memory access (CoDMA). The CoDMA ensures that the writing address is always less than the reading address in coherent read and write mode, so as to avoid read-after-write (RAW) errors. It features an efficient data exchanging scheme without using data buffer in the memory. A four-core single-instruction multiple-data processor is established for the experiments, based on a multi-bus network-on-chip. Experimental results show that the proposed method consumes no additional memory buffer and achieves 39% and 20% average performance improvement compared with traditional Methods 1 and 2, respectively. And a maximal of 43% reduction in memory usage is achieved, at the cost of only 0.22% more area overhead compared with the entire system.

Weakly labelled semi-supervised audio tagging (AT) and sound event detection (SED) have become significant in real-world applications. A popular method is teacher-student learning, making student models learn from pseudo-labels generated by teacher models from unlabelled data. To generate high-quality pseudo-labels, we propose a master-teacher-student framework trained with a dual-lead policy. Our experiments illustrate that our model outperforms the state-of-the-art model on both tasks.

A modular serial pipelined sorting architecture for continuous input sequences is presented. It supports continuous sequences, whose lengths can be dynamically changed, and does so using a very simple control strategy. It consists of identical serial cascaded sorting cells, and lends itself to high frequency implementation with any number of sorting cells, because both data and control signals are pipelined. With L cascaded sorting cells, it produces a fully sorted result for sequences whose length N is equal to or less than L+1; for longer sequences, the largest L elements are sorted out. Being modularly designed, several independent smaller sorters can be dynamically configured to form a larger sorter.

When debugging bugs, programmers often prepare test cases to reproduce buggy behaviours. However, for concurrent programs, test cases alone are typically insufficient to reproduce buggy behaviours, due to the nondeterminism of multi-threaded executions. In literature, various approaches have been proposed to reproduce buggy behaviours for concurrency bugs deterministically, but to the best of our knowledge, they are still limited. In particular, we have recognized three debugging scenarios from programming practice, but existing approaches can handle only one of the scenarios. In this paper, we propose a novel approach, called SPDebugger, that provides finer-grained thread controlling over test cases, programs under test, and even third party library code, to reproduce the predesigned thread execution schedule. The evaluation shows that SPDebugger handles more debugging scenarios than the state-of-the-art tool, called IMUnit, with similar human effort.

This paper introduces a block based motion estimation algorithm based on projection with adaptive window size selection. The blocks cannot match well if their corresponding 1D projection does not match well, with this as foundation 2D block matching problem is translated to a simpler 1D matching, which eliminates majority of potential pixel participation. This projection method is combined with adaptive window size selection in which, appropriate search window for each block is determined on the basis of motion vectors and prediction errors obtained for the previous block, which makes this novel method several times faster than exhaustive search with negligible performance degradation. Encoding QCIF size video by the proposed method results in reduction of computational complexity of motion estimation by roughly 45% and over all encoding by 23%, while maintaining image/video quality.