| Symmetric Multi Processor | Massive Parallel Processor | |
| 1 | SMP stands for Symmetric Multi processor | MPP stands for Massive parallel Processing |
| 2 | In SMP every processor share a single copy of the operating system (OS) | In MPP each processor use its own operating system (OS) and memory. |
| 3 | SMP supports shared Architecture. | MPP supports shared nothing Architecture |
| 4 | SMP is the primary parallel architecture employed in servers | MPP is the coordinated processing of a single task by multiple processors, |
| 5 | SMP architecture is a tightly coupled multiprocessor system, | In MPP each processor works on a different part of the task. |
| 6 | In SMP resources like bus, memory and an I/O system are common. | Each processor has its own set of disks |
| 7 | SMP processor share whole work between them | Each node is responsible for processing only the rows on its own disk |
| 8 | No Separate buffer pool or lock tables, All is shared | Each node maintains its own set of lock tables and buffer pool increasing usability of in memory feature |
| 9 | SMP grows by buying a bigger System | Scalability is easy by just adding racks , from few TB's to 6 peta byte |
| 10 | SMP usually faces resource contention | MPP is solution to resource contention |
| 11 | To create Distributed Architecture complex design is required and can only achieve partially. | MPP is designed to be Distributed Architecture |
| 12 | In – Memory feature provided by software is totally depending on amount of Ram and load. | Data is Horizontally Partitioned with huge compression up to 40 x explores in-memory in best manners |
| 13 | In SMP every CPU have its own cache either its dual or quad core but rest all resources are shared | MPP processors communicate between each other using some form of Messaging interface via an “Interconnect” |
Scaling


Fundamental Difference

SQL Server 2012 Parallel Data Warehouse Architecture
![]() |
| SQL Server Parallel Data Warehouse Architecture |
Recommended:What is MPP ( Massive Parallel Processor )


Very clear, concise and informative, Thank You for sharing.
ReplyDeleteWelcome !!
DeleteNow I know. thanks
ReplyDeletefix this page
ReplyDeleteIts fixed
Delete