The SQLFast-RT technique was devised in 2001 by Director Marcus Anderson. Using an Intel i586 Quad processor with 4Gb memory – or a standard desktop PC by today’s standards – this hardware configuration performed the extraction and transformation of records from 40GB of doubly compressed tape data each week in less than 1 hour. Today’s High Performance Computing centres can expect Real Time ETL capability from SQLFast-RT.
Depending on RAM and search string sizes typically only 2Gb of main memory is required for up to 7 dimension analytic cubes. This opens up the possibility today of real-time ETL on TCR’s using massively parallel blade based computing centres.
The SQLFast-RT algorithm is invulnerable to the number of search keys, making it the ideal solution for pattern matching applications that need to be search key agnostic. Whether it be 50 search keys or 50,000 search keys, SQLFast-RT will complete in the same amount of time. This capability is unheard of in typical search algorithms and Anderson Digital is not aware of any other provider of this capability.SQLFast-RT is a highly optimised SQL syntax based search engine to compress its binary searches down to only a few machine cycles for each record, regardless of the search key or the number of search keys.
SQLFast-RT is particularly useful in Commercial, Forensic, Military, and Authorised Surveillance applications where legitimate access to metadata records is available but massive data sizes prevent useful analysis.
SQLFast-RT can also be adapted for high speed MDAC applications that involve Big Data or massive volumes of data, even, we claim, as voluminous as the phenomenal amounts of data accumulated by the The Large Hadron Collider (LHC).
Interested? Please contact Marcus Anderson
No comments:
Post a Comment