Sparse matrix–matrix multiplication (SpMM) is a fundamental operation in scientific computing with broad applications across numerous domains. Tiling is a key optimization technique for improving data locality and is widely adopted in high-performance computing. However, the irregular data access patterns inherent to SpMM make it challenging to exploit tiling effectively for data reuse. In this article, we propose MaSpMM , a memory-aware SpMM framework that integrates cache-aware tiling with a segment-oriented data layout. MaSpMM stores matrices as continuous segments to enhance data locality within each tile. Moreover, since many sparse matrices in real-world applications exhibit symmetry, we further develop MaSpMM-Sym, an extension that recursively partitions symmetric matrices to eliminate write conflicts and further improve locality. To adapt to diverse scenarios, we finally introduce MaSpMM-Adap, which adaptively selects the most suitable approach for each input matrix. Comprehensive evaluations on both x86 and ARM CPUs demonstrate that MaSpMM-Adap achieves average speedups of up to 1.86× over Intel oneMKL, 1.84× over ASpT, and 1.75× over J-Stream.

We present a regularized inversion method for three-dimensional (3D) magnetotelluric (MT) data with axial anisotropic conductivities based on the edge-based finite element (FE) method. The Gauss–Newton (GN) approach is used to minimize the inversion objective function, including data misfit and regularization penalties, considering both structural complexity and anisotropic penalties. The most time-intensive task in the 3D MT inversion process is solving the large sparse system of linear equations. To speed up the inversion calculation, a hybrid direct–iterative solver combined with a block-diagonal preconditioner that has not yet been applied in anisotropic inversion is developed to accelerate the solutions for the sparse linear system resulting from forward modeling and sensitivity computations. In each GN iteration, a preconditioned conjugate gradient (PCG) method is adopted to overcome the difficulty of the sensitivity matrix storage for the anisotropic scene and obtain a model update without explicitly calculating and storing the sensitivity matrix. Before the inversion test, we use a model to demonstrate that the hybrid solver is computationally beneficial in terms of memory usage and time spent as compared to the direct solver. The good convergence properties and efficiency of the GN–PCG inversion scheme are demonstrated by two synthetic models and USArray data. The proposed inversion scheme can be an important supplement to existing anisotropic inversion algorithms and provide technical support for MT data interpretation.

As intelligent transportation systems (ITS) are now being integrated into our everyday lives, it has been widely accepted that forecasting road networks is a promising killer engine for ITS with high social and economic benefits. However, current solutions ignore the heterogeneity of spatial-temporal traffic data and fail to capture hidden spatial-temporal correlations. This paper presents STEGNN: a novel spatial-temporal embedding graph neural network for road network forecasting. The key idea of STEGNN is utilizing Cosine Similarity to generate a high-quality temporal graph and thus fills the gap between the temporal-spatial correlations for traffic graph, which includes (i) a novel approach to construct temporal graph based on temporal-spatial similarity from traffic graphs, which is much more accurate on measured similarity of time series claimed by previous methods; (ii) an advanced spatial-temporal embedding model to exploit spatial-temporal dependencies by leveraging specific arrangements of temporal and spatial graphs; and (iii) an effective framework that gasps extensive spatial-temporal dependencies in the long-term by mixing multi-layer graph convolution with dilated convolution to understand wide-range spatial-temporal features. Extensive evaluations validate STEGNN by applying it to real-world traffic graphs and indicate that STEGNN outperforms state-of-the-art solutions with much more accurate forecasting of road networks.

As the era of exascale supercomputing is coming, it is vital for next-generation supercomputers to find appropriate applications with high social and economic benefit. In recent years, it has been widely accepted that extremely-large graph computation is a promising killer application for supercomputing. Although Tianhe series supercomputers are leading in the world-wide competition of supercomputing (ranked No. 1 in the Top500 list for six times), previously they had been inefficient in graph computation according to the Graph500 list. This is mainly because the previous graph processing system cannot leverage the advanced hardware features of Tianhe supercomputers. To address the problem, in this paper we present our integrated optimizations for improving the graph computation performance on our next-generation Tianhe supercomputing system, mainly including sorting with buffering for heavy vertices, vectorized searching with SVE (Scalable Vector Extension) on matrix2000+ CPUs, and group communication on the proprietary interconnection network. Performance evaluation on a subset of the Tianhe supercomputer (with 512 nodes and 196,608 cores) shows that our customized graph processing system effectively improves the graph search performance and achieves the BFS performance of 2131.98 GTEPS.