In the Diver Platform, Spectre is the data analysis software. Spectre is constructed to take advantage of the latest hardware advances, such as parallel processing, larger amounts of memory, and advanced compiler technology.
Spectre processes data from one or more databases and files to build a column-oriented database (cBase) that caches efficiently on both the server and client device. This read-only columnar database is designed for efficient aggregation, while easily ignoring columns that are not needed. This flexibility reduces the need to make up-front decisions about data organization. Features in Spectre Build can assist with data assessment and validation.
Spectre is used not only to build cBases—it also powers the DiveLine server software for efficient queries from DI clients against those cBases. Spectre includes a calculation engine that compiles formulas into machine code to maximize hardware utilization. The result is efficient real-time calculations for time series, dimension counts, and multitabs.
Spectre can be used as a database platform: the place to not only load the data, but to perform additional ETL work. Think of a cBase as a table within the database. Anything you can do with a database table you can do with a cBase. Building and diving are very similar operations using Spectre. You can treat Spectre as a work environment that you access and manipulate with Workbench.
For more information on the Spectre language, see:
- Hardware and OS Requirements
- Scripting Syntax Elements
- Scripting Basics: SDL Syntax
- Spectre Data Types
- Spectre Memory Usage