In a post on parallel backup in MySQL Enterprise Backup (MEB) v3.8.0 jothir (Hema Sridharan) explains how read/write operations in this version are now performed in parallel using pipeline parallelism. The post explains that the pipelined parallel execution of read, process (if requested) and write operations is achieved by invoking multiple threads for different operations and usage of buffers. Each operation is split into segments, and threads operate on different segments from different iterations concurrently. Each segment is assigned one or more worker threads and in-queue buffers which stores the work to be processed by that stage. The post provides an example of these operations.