MPP

(redirected from Massively parallel)
Also found in: Dictionary, Acronyms, Wikipedia.
Related to Massively parallel: Massively parallel computing

MPP

Multiple Parallel Processing

MPP

(Massively Parallel Processing or Massively Parallel Processor) A multiprocessing architecture that uses many processors and a different programming paradigm than the common symmetric multiprocessing (SMP) found in today's computer systems.

Self-Contained MPP Subsystems
Each CPU is a subsystem with its own memory and copy of the operating system and application, and each subsystem communicates with the others via a high-speed interconnect. In order to use MPP effectively, an information processing problem must be breakable into pieces that can all be solved simultaneously. In scientific environments, certain simulations and mathematical problems can be split apart and each part processed at the same time. In the business world, a parallel data query (PDQ) divides a large database into pieces. For example, 26 CPUs could be used to perform a sequential search, each one searching one letter of the alphabet. To take advantage of more CPUs, the data have to be broken further into more parallel groups.

In contrast, adding CPUs in an SMP system increases performance in a more general manner. Applications that support parallel operations (multithreading) immediately take advantage of SMP, but performance gains are available to all applications simply because there are more processors. For example, four CPUs can be running four different applications. See SMP.


MPP and SMP Architecture
In MPP operation, the problem is broken up into separate pieces, which are processed simultaneously. In SMP, CPUs are assigned to the next available task or thread that can run concurrently.


MPP and SMP Architecture
In MPP operation, the problem is broken up into separate pieces, which are processed simultaneously. In SMP, CPUs are assigned to the next available task or thread that can run concurrently.







Integrated MPP and SMP
In the mid-1990s, Pyramid Technology combined MPP and SMP processing in its Reliant line. The SMP systems attached to a high-speed mesh interconnect that provided unusual flexibility, and the architecture was passed on to future systems after Fujitsu acquired the company. See Fujitsu Siemens. (Image courtesy of Pyramid Technology Corporation.)







ServerNet Mesh Interconnect
The orange disks, purple CPUs and green channels conceptually show the interconnections between CPU and I/O subsystems. ServerNet was a high-speed switch from Tandem when it was a thriving company. (Image courtesy of Tandem Computers.)
References in periodicals archive ?
A massively parallel switch router is ideally suited to leverage its excess processing power to provide these services at wire-speed, without the need for additional hardware.
The massively parallel, shared-nothing set of SPUs provides the performance advantage of MPP.
We are seeing the arrival of smaller machines, massively parallel machines, machines with new architectures designed for ease of software development rather than raw number-crunching capability.
7 million contract from the Defense Advanced Research Projects Agency to develop a new, massively parallel computer system that can execute 1 billion instructions per second.
These universities, and their distinguished professors and graduate students, are extending the boundaries of knowledge in massively parallel and embedded computing.
We continue to expand our intellectual property position and establish MPT as an innovator in the field of parallel processing," noted Scott Smith, chief executive officer of Massively Parallel Technologies.
4 provides support for Massively Parallel Processing, allowing organizations to distribute large data integration jobs across multiple servers to maximize throughput and minimize batch windows when updating large data warehouses.
I first learned about Massively Parallel Technologies about a year ago and found it to be a brilliant and efficient system, perhaps the most efficient and scalable system one can find," noted Duncan.
Kilocore Massively Parallel Architecture features unprecedented performance per watt while running multiple applications in parallel.
18 /PRNewswire/ -- Massively Parallel Technologies (MPT), a provider of on-demand high-performance computing, announced the company has developed the first mathematical derivation of Amdahl's Law.
Low Latency Switch Allows Quadrics to Expand Its Line of 10-Gigabit Ethernet Interconnects for Massively Parallel Computing
6 /PRNewswire/ -- Massively Parallel Technologies (MPT), a provider of on-demand high-performance computing, recently hosted an event at MPT headquarters featuring recognized authorities on parallel processing Dr.

Full browser ?