This article explains the underlying concepts of Teradata engine architecture. All editions of Teradata use the same engine.
Teradata's architecture is based on a Massively Parallel Processing (MPP), shared-nothing architecture that enables high-performance data processing and analytics. MPP distributes workloads across multiple vprocs, or virtual processors. The virtual processor responsible for query processing is commonly referred to as Access Module Processor (AMP). Each AMP is isolated from other AMPs, and processes queries in parallel, allowing Teradata to process large volumes of data rapidly.
The major architectural components of the Teradata engine include the Parsing Engines (PEs), BYNET, Access Module Processors (AMPs), and Virtual Disks (Vdisks). Vdisks are assigned to AMPs in enterprise platforms.