Adaptive unified memory management method and system for large-scale graphs

An adaptive large graph-oriented unified memory management method, characterized in that, an ordered setting procedure according to a priority of different types of graph data in a graph computing application, and a policy making procedure based on whether a GPU memory is full and a relative size of targeted data and available GPU memory capacity, and a described unified memory management strategy configuration sets the management strategy of the current graph data by calling cudaMemPrefetchAsync and cudaMemAdvise instructions, and data usage hints include AccessedBy and ReadMostly, wherein the cudaMemPrefetchAsync moves part of data to the GPU memory, the cudaMemAdvise is set for a memory usage hint that guides a GPU driver to control data movement for improving performance, the AccessedBy and ReadMostly are instructions for a series of GPUs, and the described unified memory management policy configuration specifically includes:

• (i) for Vertex Property data, when the GPU memory is full, set a hint of Vertex Property to AccessedBy;
• otherwise, that is, when the GPU memory is not full and when the Vertex Property does not exceed the GPU's available memory capacity, set a prefetching size of Vertex Property to the size of the Vertex Property;
• when the Vertex Property exceeds the available memory capacity of the GPU, set the hint of the Vertex Property to AccessedBy, and set a prefetching amount of the Vertex Property as:
• prefetch rate×available memory capacity of the GPU, measured in bytes;
• (ii) for Vertex Offset data, when the GPU memory is full, set the hint of Vertex Offset to AccessedBy;
• otherwise, that is, when the GPU memory is not full and when the Vertex Offset does not exceed the GPU's available memory capacity, set the prefetching size of Vertex Offset to the size of the Vertex Offset;
• when the Vertex Offset exceeds the available memory capacity of the GPU, set the hint of the Vertex Offset to AccessedBy, and set the prefetching amount of the Vertex Offset as:
• prefetch rate×available memory capacity of the GPU, measured in bytes;
• for Frontier data, when the GPU memory is full, set the hint of Frontier to AccessedBy;
• otherwise, that is, when the GPU memory is not full and when the Frontier does not exceed the GPU's available memory capacity, set the prefetching size of Frontier to the size of the Frontier;
• when the Frontier exceeds the available memory capacity of the GPU, set the hint of the Frontier to AccessedBy, and set the prefetching amount of the Frontier as:
• prefetch rate×available memory capacity of the GPU, measured in bytes;
• for Edge data, when the GPU memory is full, set the hint of Edge to AccessedBy;
• otherwise, that is, when the GPU memory is not full and when the Edge does not exceed the GPU's available memory capacity, set the prefetching size of Edge to the size of the Edge;
• when the Edge exceeds the available memory capacity of the GPU, set the hint of the Edge to AccessedBy, and set the prefetching amount of the Edge as:
• prefetch rate×available memory capacity of the GPU, measured in bytes.