With the advent of Big Data sets mainstream technologies like Massively Parallel Processing (MPP) systems is experiencing vital growth. Let’s discuss about what is MPP or Massively Parallel Processing (MPP)
Massively parallel processing (MPP) has been designed mainly for business intelligence and analytical processing. Massively Parallel Processing (MPP) architecture consists of multiple servers with each server or node autonomous to process and store up data. These multiple severs or processors executes in parallel to provide high performance. MPP is alike to symmetric processing (SMP), with the only difference that in MPP systems each CPU has its own memory whereas in SMP systems all the CPUs share the same memory. MPP systems don’t go through bottleneck problem like SMP systems in which all CPUs attempt to access same memory at once. MPP systems avoid this bottleneck by distributing data and processing across many servers (nodes) each of which has its own memory and disk thus sharing the load.
Massively Parallel Processing is in general used in applications like DW appliances, decision support systems and data warehouses. High volumes of data in data warehouse systems are made into smaller and more manageable blocks, which are then distributed to multiple processors. Also there are no disk-level sharing, all communication is through network interconnect.
MPP systems are used by large companies due to their high cost and complexity. Some of the favorite MPP databases are Teradata,Greenplum and Netezza.We will discuss about these databases in upcoming posts.
If you like this post, please share it on google by clicking on the Google +1 button.
Please go through our latest post TOP 6 BIG DATA TRENDS IN THE NEAR FUTURE