High Dimensional Massive Data Processing Based on Locality-Sensitive Hashing and Double-Layer Skiplist

When processing the high-dimension and massive data collected in the production process of ultra-precision machining tool, there is the issue of low efficiency in query operation of multi-key query and time range. For solving this issue, this paper propose a high-dimensional and massive data storage query method based on locality-sensitive hashing algorithm and double-layer skiplist index structure to improve the efficiency of multi-key query and time range query. In this method, by using the locality-sensitive hashing algorithm, the machine state parameters are stored in multiple similar data nodes, and the double-layer skiplist index structure is established in each data node. This method reduces the search range of multi-key query operations at the cost of a certain probability error, thereby indirectly improving query efficiency. Experimental results show that the method is effective.