Amazon Redshift is the data warehousing service that forms a key part of the overall AWS cloud solution. Redshift enables organisations, both large and small, to store many different types of data as a central cloud-based repository and keep it ready for querying. For example, to spot customer behaviour trends amongst multiple customer segments in industries such as in retail, banking, insurance or telecoms, enabling an organisation to explore several different data sources and maximise the insights that this data can bring.
Redshift has commonality with relational databases as it stores its data in a structured way, however, a key difference that Redshift has versus conventional databases is that it uses a column-orientated structure as opposed to row-orientated. Storing database tables in this columnar structure drastically reduces the overall requirements on disk I/O, thereby reducing the amounts of data that are being loaded from the disk. This is a key factor in enabling Redshift to optimize analytics performance.
In essence, the column-orientated structure means that each data block stores values of a single column for multiple rows of data. As data is loaded onto the Redshift data warehouse, it is automatically converted to the columnar structure.
This structure enables Redshift to use a compression scheme which is optimised for each particular data type, providing disk space savings and efficiencies. These space savings in turn enable more data to be loaded directly into memory as analytic work is being carried out, which reduces the requirements on disk I/O.
Boosted performance
With the demands that are being put on analytics today, by the sheer growth of data sets and the number of data points that are available to businesses, ever more efficiencies are required from data warehouses to meet the demands that are being put upon it and maximise the insights and competitive advantages that can be drawn out.
In order to unleash the maximum power and flexibility of Amazon Redshift, Brytlyt brings speed of thought processing, with the world’s fastest GPU data processing engine. This enables analytics and BI visualisation tools, such as Tableau, Power BI and TIBCO Spotfire, to be pushed to their limits, without the need for costly and time-consuming pre-loading or pre-aggregating of data sets. Data pre-aggregation projects in themselves can take many months of specification, planning and development, thereby consuming valuable resource, and potentially narrowing the data being made available to end-users and consumers of such projects.
Read our fast data loading blog for more or get in touch to learn about how Brytlyt’s GPU database can accelerate your Redshift solution.
