 |
|
|
|
|
|
| |
DATAllegro Data Warehouse Appliances
DATAllegro v3™ goes beyond the low cost and high performance of first generation data warehouse appliances and adds the flexibility and scalability that only an enterprise-class platform can offer. The result is a complete data warehouse appliance that enables companies to rapidly query large volumes of data (up to several hundreds of terabytes) at a very affordable price.
HOW DOES IT WORK?
DATAllegro's architecture was designed to be flexible, modular and open in order to leverage the most reliable and advanced, standards-based technologies on the market. With DATAllegro v3, DATAllegro has achieved its goal of an open, flexible platform. V3 utilizes EMC® storage, Dell™ servers, Cisco® InfiniBand switches, Intel® multi-core CPUs and the Ingres® open source database. The result is the low cost and high performance from first generation appliances as well as the reliability and scalability of second generation appliances.
|
|
RAIDW™
The use of traditional (typically SAN-based) storage solutions in data warehouse environments slows the speeds at which data can be queried, and reduces the amount of disk space available for user data. DATAllegro’s patent-pending RAIDW™ technology maximizes data I/O speeds, while also providing built-in fault tolerance for all components, not just the disks.
DIRECT DATA STREAMING™ (DDS)
Most traditional data warehouse technologies end up with random I/O at the disk level. This is due to the extensive use of indexes, complex SAN disk infrastructures and the need to support multiple hardware platforms and operating systems. DATAllegro’s Direct Data Streaming™ technology enables most of your queries to access data using sequential disk I/O without tuning.
ULTRA-SHARED NOTHING™ (USN) PARALLEL DATABASE
DATAllegro’s USN™ parallel database design maximizes the use of co-located joins across all nodes. The result is reduced network traffic which means even faster query speeds for all query types. Multi-level partitioning minimizes the amount of data read for each query to improve performance even further. Finally, indexes can be used to give outstanding performance on queries that only require a few rows of data.
|
|
| |
|
|
|
| |
|
