Semiconductor test data processing method and device

Abstract

The invention provides a semiconductor test data processing method, which comprises the following steps of: reading test data with the same device name under different wafers or crystal grains in batches from a database, and determining the test data corresponding to the device name; determining a selected device name according to a user demand, extracting test data corresponding to the selected device name, and preprocessing the test data; and grouping the extracted test data, determining a grouping number based on a specific quantile, and obtaining a quantile corresponding to the specific quantile based on the minimum value of the group corresponding to the adjacent grouping number and the corresponding quantile thereof. Effective quantile calculation and data statistics can be carried out under limited resource configuration, the analysis complexity is simplified, and the efficiency is high. The invention also provides a semiconductor data processing device, which can realize the data processing method provided by the invention, has corresponding advantages, can better realize accurate calculation of quantiles, and is beneficial to improvement of production efficiency.

Description

technical field [0001] The invention belongs to the technical field of data processing and analysis, and in particular relates to a data processing method and a data processing device involved in semiconductor testing. Background technique [0002] With the continuous improvement of semiconductor process technology and the improvement of testing capabilities, both design companies and production foundries have a large amount of data related to semiconductor testing that needs to be processed and analyzed, the most prominent of which is testing during wafer testing. The number is increasing, and the test data that needs to be analyzed and processed is getting larger and larger. When a large amount of test data pours into the data warehouse, data analysts face such a huge test data, how to perform fast and effective processing and analysis, especially how to perform effective quantile calculation and data statistics on the grouped data, become increasingly urgent. For exampl...

AI Technical Summary

Problems solved by technology

Therefore, even if an effective distributed computing platform is built, a large number of servers are needed to build a platform data analysis platform if one wants to quickly obtain calculation results. In addition, for the sake of security and data confidentiality, we cannot rely entirely on building a data analysis platform to process the data involved in semiconductor testing.

Images