Content deleted Content added
m Punctuation |
DAZMasters (talk | contribs) |
||
Line 26:
Data warehouse appliances architecturally integrate database, server and storage components into a single unit.<ref name="CRN">[http://www.crn.com/news/components-peripherals/210603775/oracle-with-hp-debuts-database-server-storage-hardware-systems.htm "Oracle, With HP, Debuts Database Server, Storage Hardware Systems," "CRN," September 24, 2008]</ref> Netezza's appliances use a proprietary Asymmetric [[Massively Parallel Processing]] (AMPP) architecture that combines open, blade-based servers and disk storage with a proprietary data filtering process using [[field-programmable gate array]]s (FPGAs).<ref name="Register2">[http://www.theregister.co.uk/2010/02/24/netezza_data_analytics/ "Netezza to bake analytics into appliances," "The Register," February 24, 2010]</ref>
FPGAs are increasingly used in conventional high performance computing applications where specific computational kernels are performed on the FPGA instead of a microprocessor. Netezza’s FPGA implementation performs complex data filtering so that only relevant portions of [[big data]] sets are passed along to the processor to run the [[SQL]] query. Additionally, spare FPGA cycles are used to perform data compression and decompression, reducing the load on the processor. Because the FPGA handles the compression, and the data set is smaller, data throughput is accelerated. [[In-database processing]] reduces latency and speeds data analysis by eliminating the need to move data out of and results back into the warehouse.<ref name="Register2" />.
==History==
| |||