This paper proposes a framework called SnuRHAC, which provides an illusion of a single GPU for the multiple GPUs in a cluster. Under SnuRHAC, a CUDA program designed to use a single GPU can utilize multiple GPUs in a cluster without any source code modification. SnuRHAC automatically distributes workload to multiple GPUs in a cluster and manages data across the nodes. To manage data efficiently, SnuRHAC extends CUDA Unified Memory and exploits its page fault mechanism. We also propose two prefetching techniques to fully exploit UM and to maximize performance. Static prefetching allows SnuRHAC to prefetch data by statically analyzing CUDA kernels. Dynamic prefetching complements static prefetching. SnuRHAC enforces an application to run on a single GPU if it is not suitable for multiple GPUs. We evaluate the performance of SnuRHAC using 18 benchmark applications from various sources. The evaluation result shows that while SnuRHAC significantly improves ease-of-programming, it shows scalable performance for the cluster environment depending on the application characteristics.

We study a large-scale resource allocation problem with a convex, separable, not necessarily differentiable objective function that includes uncertain parameters falling under an interval uncertainty set, considering a set of deterministic constraints. We devise an exact algorithm to solve the minimax regret formulation of this problem, which is NP-hard, and we show that the proposed Benders-type decomposition algorithm converges to an [Formula: see text]-optimal solution in finite time. We evaluate the performance of the proposed algorithm via an extensive computational study, and our results show that the proposed algorithm provides efficient solutions to large-scale problems, especially when the objective function is differentiable. Although the computation time takes longer for problems with nondifferentiable objective functions as expected, we show that good quality, near-optimal solutions can be achieved in shorter runtimes by using our exact approach. We also develop two heuristic approaches, which are partially based on our exact algorithm, and show that the merit of the proposed exact approach lies in both providing an [Formula: see text]-optimal solution and providing good quality near-optimal solutions by laying the foundation for efficient heuristic approaches.

Temporal data are prevalent, where one or several time attributes present. It is challenging to identify the temporal attributes from heterogeneous sources. The reason is that the same attribute could contain distinct values in different time spans, whereas different attributes may have highly similar timestamps and alike values. Existing studies on schema matching seldom explore the temporal information for matching attributes. In this paper, we argue to order the values in an attribute A by some time attribute T as a time series. To learn deep temporal features in the attribute pair (T, A), we devise an auto-encoder to embed the transitions of values in the time series into a vector. The temporal attribute matching (TAM) is thus to evaluate matching distance of two temporal attribute pairs by comparing their transition vectors. We show that computing the optimal matching distance is NP-hard, and present an approximation algorithm. Experiments on real datasets demonstrate the superiority of our proposal in matching temporal attributes compared to the generic schema matching approaches.

In this paper, we propose a runtime, called HUM, which hides host-to-device memory copy time without any code modification. It overlaps the host-to-device memory copy with host computation or CUDA kernel computation by exploiting Unified Memory and fault mechanisms. HUM provides wrapper functions of CUDA commands and executes host-to-device memory copy commands in an asynchronous manner. We also propose two runtime techniques. One checks if it is correct to make the synchronous host-to-device memory copy command asynchronous. If not, HUM makes the host computation or the kernel computation wait until the memory copy completes. The other subdivides consecutive host-to-device memory copy commands into smaller memory copy requests and schedules the requests from different commands in a round-robin manner. As a result, the kernel execution can be scheduled as early as possible to maximize the overlap. We evaluate HUM using 51 applications from Parboil, Rodinia, and CUDA Code Samples and compare their performance under HUM with that of hand-optimized implementations. The evaluation result shows that executing the applications under HUM is, on average, 1.21 times faster than executing them under original CUDA. The speedup is comparable to the average speedup 1.22 of the hand-optimized implementations for Unified Memory.

Structured entities are commonly abstracted, such as from XML, RDF or hidden-web databases. Direct retrieval of various structured entities is highly demanded in data lakes, e.g., given a JSON object, to find the XML entities that denote the same real-world object. Existing approaches on evaluating structured entity similarity emphasize too much the structural inconsistency. Indeed, entities from heterogeneous sources could have very distinct structures, owing to various information representation conventions. We argue that the retrieval could be more tolerant to structural differences and focus more on the contents of the entities. In this paper, we first identify the unique challenge of parent-child (containment) relationships among structured entities, which unfortunately prevent the retrieval of proper entities (returning parents or children). To solve the problem, a novel hierarchy smooth function is proposed to combine the term scores in different nodes of a structured entity. Entities sharing the same structure, namely an entity family, are employed to learn the coefficient in aggregating the scores, and thus distinguish/prune the parent or child entities. Remarkably, the proposed method could cooperate with both the bag-of-words (BOW) and word embedding models, successful in retrieving unstructured documents, for querying structured entities. Extensive experiments on real datasets demonstrate that our proposal is effective and efficient.

Extensive prior studies aimed at the development of diagnostic methods for planetary gearboxes have mainly examined acceleration and acoustic emission signals. Recently, due to the relationship between gear mesh stiffness and transmission error (TE), TE-based techniques have emerged as a promising way to analyze dynamic behavior of spur and helical gears. However, to date, TE has not been used as a measure to detect faults in planetary gears. In this paper, we propose a new methodology for model-based fault diagnostics of planetary gears using TE signals. A lumped parametric model of planetary gear dynamics was built to extract simulated TE signals, while accounting for the planet phasing effect, which is a peculiar characteristic of the planetary gear. Next, gear dynamic analysis was performed using TE signals, and TE-based damage features were calculated from the processed TE signals to quantitatively represent health conditions. The procedures described aforesaid were then applied to a case study of a planetary gear in a wind turbine gear train. From the results, we conclude that TE signals can be used to detect the faults, while enhancing understanding of the complex dynamic behaviors of planetary gears.