Matrix computation, in particular, matrix multiplication is time-consuming, but essentially and widely used in a large number of applications in science and industry. The existing distributed matrix multiplication methods only focus on either low communication cost (i.e., high performance) with the risk of out of memory or large-scale processing with high communication overhead. We propose a distributed elastic matrix multiplication method called CuboidMM that achieves both high performance and large-scale processing. We also propose a GPU acceleration method that can be combined with CuboidMM. CuboidMM partitions matrices into cuboids for optimizing the network communication cost with considering memory usage per task, and the GPU acceleration method partitions a cuboid into subcuboids for optimizing the PCI-E communication cost with considering GPU memory usage. We implement a fast and elastic matrix computation engine called DistME by integrating CuboidMM with GPU acceleration on top of Apache Spark. Through extensive experiments, we have demonstrated that CuboidMM and DistME significantly outperform the state-of-the-art methods and systems, respectively, in terms of both performance and data size.

BACKGROUND: A cross-correlation (XCorr) score function is one of the most popular score functions utilized to search peptide identifications in databases, and many computer programs, such as SEQUEST, Comet, and Tide, currently use this score function. Recently, the HiXCorr algorithm was developed to speed up this score function for high-resolution spectra by improving the preprocessing step of the tandem mass spectra. However, despite the development of the HiXCorr algorithm, the score function is still slow because candidate peptides increase when post-translational modifications (PTMs) are considered in the search. RESULTS: We used a graphics processing unit (GPU) to develop the accelerating score function derived by combining Tide's XCorr score function and the HiXCorr algorithm. Our method is 2.7 and 5.8 times faster than the original Tide and Tide-Hi, respectively, for 50 Da precursor tolerance. Our GPU-based method produced identical scores as did the CPU-based Tide and Tide-Hi. CONCLUSION: We propose the accelerating score function to search modifications using a single GPU. The software is available at https://github.com/Tide-for-PTM-search/Tide-for-PTM-search .

Satellite image data generated from remote sensors around the world have different resolutions and are processed at varying levels from Level 0 to Level 3, with each level containing vast amounts of information. Due to the problem of data size, many ocean science researchers use L3 images, which have a spatial resolution of 4 km or 9 km. However, in order to overcome problems such as red tides or to analyze the marine ecosystem based on ocean color satellite research, researchers must generate data by changing the parameters of images at various levels. There is also a need for immediate access to satellite image information using analytical and visualization tools. Considering those requirements, TuPiX-OC (Turning PiXels into knowledge and science-OceanColor) provides an environment to design and execute algorithms for data processing and analysis of satellite image data by data type. TuPix-OC, which has a distributed architecture, is an analytical platform that supports data import, level conversion, DB integration, analysis and processing, and visualization. TuPiX-OC stores satellite data in a massive storage device, and provides an online platform for satellite data conversion/analysis/ visualization. For satellite data processing, TuPiX-OC converts NASA-provided binary files into files that can be analyzed by users. Moreover, TuPiX-OC includes various algorithms for satellite data selection and utilization of satellite images. Preliminary Experiments of TuPiX-OC's satellite image data processing capability showed that it was able to process 35 times as many images as the open source software SeaDAS.