Barrier synchronization, an essential mechanism for a block of threads to guard data consistency, is regarded as a threat to performance. This study, however, provides a different viewpoint for barrier synchronization on GPUs: adding barrier synchronization, even when functionally unnecessary, can improve the performance of some memory-intensive applications. We explain this phenomenon using a memory contention model in which artificial barrier synchronization helps reduce memory contention and preserve data access locality. To yield practical applications, we identify a program pattern: artificial barrier synchronization can be used to synchronize the memory accesses when the data locality among threads is violated. Empirical results from three real-world applications demonstrate that artificial barrier synchronization can increase performance by 10 to 20 percent.

As the computational power of Graphics Processing Unit (GPU) increases, data transmission becomes the major performance bottleneck. In this study, we investigate two techniques, data streaming and data compression, to reduce the communication cost on GPU. Data streaming enables overlap of communication and computation, whereas data compression reduces the data size transferred among different memory spaces. Although both techniques increase computation cost, overall performance can still be enhanced by reducing communication cost. We demonstrate the effectiveness of the two techniques via two case studies: radix sort and 3-star, a deployment algorithm in wireless sensor networks. For radix sort, a new algorithm, which mixes MSD and LSD algorithms and employs data streaming, is presented. Its performance is 25% faster than the fastest GPU radix sort implementation currently available in the public domain. For the 3-star algorithm, the speed increases several hundreds of times faster than that obtained by the CPU code. The data streaming and data compression, which is a hybrid CPU-GPU algorithm, provide an additional 54% performance improvement to the GPU implementation. Data compression not only reduces communication cost, but also improves the computation time, by which further performance enhancement can be achieved.

In this paper, we investigate efficient algorithms and implementations using GPU plus CPU to solve the rectangle intersection problem on a plane. The problem is to report all intersecting pairs of iso-oriented rectangles, whose parallelization on GPUs poses two major computational challenges: data partition and the massive output. The algorithm we presented is called PRI-GC, Parallel Rectangle Intersection algorithm on GPU+CPU, which consists of two phases: mapping and intersection-checking. In the mapping phase, rectangles are hashed into different subspaces (called cells) to reduce the unnecessary intersection checking for far-apart rectangles. In the intersection-checking phase, pairs of rectangles within the same cell are examined in parallel, and the intersecting pairs of rectangles are reported. Several optimization techniques, including rectangles re-ordering, output data compressing/encoding, and the execution overlapping of GPU and CPU, are applied to enhance the performance. We had evaluated the performance of PRI-GC and the result shows over 30x speedup against two well-implemented sequential algorithms on single CPU. The effectiveness of each optimization technique for this problem was evaluated as well. Several parameters, including different degrees of rectangle coverage, different block sizes, and different cell sizes, were also experimented to explore their influences on the performance of PRI-GC.

A geotagged photo refers to that a photo contains its location information where it was taken. Compared with other existing applications, we show the photo clusters instead of all photos where the number of photo clusters is much less than that of all photos. We designed a clustering algorithm that clusters geotagged photos (abbreviated as GPCA) in accordance to thresholds of different scales. In general, GPCA is an alternative of k-means clustering algorithm. Given a scale graduation of a map, the distance threshold is also determined. Before clustering geotagged photos, we use the distance threshold to decide the number of clusters. In other words, the number of centroids is chosen according to the graduation scale of a map. GPCA clusters geotagged photos into different photo clusters. Then,GPCA recomputed the new centeriods of photo clusters. GPCA executes the computation of new centeriods iteratively until all centeriods are fixed.Note that prior partition-based clustering algorithms choose an arbitrary number of initial centriods at random. Also, choosing the number of centriods is considered as an application-dependent issue. As a result, they did not guarantee that the clustering quality is not the same every time. By taking advantage of the result of using photo clusters, the performance of map-enabled photo service is improved efficiently.