Principle of everything for fast search

One 、Everything The principle of fast search ：

Everything Does not scan the entire disk , Just reading from the disk USN journal , And index it , Store in the database in advance , Search in the database when searching , So fast .

Two 、Everything Its shortcomings and shortcomings are also obvious ：

(1) Only support NTFS Format partition , because USN The journal is NTFS Proprietary . stay FAT、FAT32 Cannot be used on format partition Everything.  (2) Index only file names 、 Date and size , Do not index file contents and additional attributes .     according to Everything According to the official website of , it 1 Minutes can index 100 Ten thousand documents .48 More than 10000 documents , It takes only a few seconds to build an index , The index does not need to scan the hard disk files one by one , It's a direct read NTFS File system USN journal ,Everything Because the core principle is based on NTFS On the underlying mechanism ,NTFS In the file system USN The log records the system's response to NTFS All changes made to files in the partition . For each volume ,NTFS All use USN Log to track the addition of 、 Delete and modify file information . Direct reading NTFS File system USN journal , So it's very fast Everything The data is stored in Everything.db In the database . With more files , The larger the index , The larger the database . You can also delete it to rebuild the index .