No abstract

The Architecture Engineering and Construction/ Facility Management (AEC/FM) industry plays a significant role in the development of economy. In recent years, the wide application and development of Building Information Modelling (BIM) promote the development of informatization and digitalization of AEC/FM industry. However, due to the limitation of a one single tool and lack of understanding of single source of truth, the problems of the industry cannot be solved completely. Therefore, the revolution and innovation of industry can be stagnated. Internet of Thing (IoT) and blockchain can be considered as two technologies that can be integrated with BIM for AEC/FM industry. The aim of this paper is to understand and analyse the basic principles and the applications of these three technologies in AEC/FM industry through literature review. With the integration of these three technologies, the virtual and realistic object and data during the whole building lifecycle can be managed and stored in a security, transparency and convenient decentralized common data environment (DCDE). Finally, a theory named, Cup-of-Water theory is presented. Keywords-BIM IoT blockchain AEC/FM industry whole building lifecycle

No abstract

No abstract

Graph neural networks (GNNs) are gaining popularity as a promising approach to machine learning on graphs. Unlike traditional graph workloads where each vertex/edge is associated with a scalar, GNNs attach a feature tensor to each vertex/edge. This additional feature dimension, along with consequently more complex vertex-and edge-wise computations, has enormous implications on locality and parallelism, which existing graph processing systems fail to exploit.This paper proposes FeatGraph to accelerate GNN workloads by co-optimizing graph traversal and feature dimension computation. FeatGraph provides a flexible programming interface to express diverse GNN models by composing coarse-grained sparse templates with fine-grained user-defined functions (UDFs) on each vertex/edge. FeatGraph incorporates optimizations for graph traversal into the sparse templates and allows users to specify optimizations for UDFs with a feature dimension schedule (FDS). FeatGraph speeds up end-to-end GNN training and inference by up to 32× on CPU and 7× on GPU.