Many competing database solutions try to solve the problem of performance by brute force, such as by throwing as much compute as possible at the problem. This works to a point, but runs into issues with unpreditable and spiraling costs when scaling. After all, compute resources aren't free!
Instead, we address this by indexing the data to maximise scanning effiency, so that even if you only deploy a few machines, those few machines can spend their compute resources scanning the data that is likely to be relevant to the query result that needs to be retrieved.
This means faster results requiring less compute power = more cost-efficient querying that can be scaled out.
Almost all enterprise data analytics platforms suffer from data silos and a lack of flexibility. Often, each new application requires a new database instance and a separate data copy to be made, massively increasing storage costs, making the whole system complex and introducing issues such as synchronisation overheads. This combination of added cost and complexity creates barriers to new application development, discourages new innovation and limits the value potential of data within your organisation.
GeoSpock DB follows a concept of dynamic data fusion – a new design philosophy which follows a single-instance-many-applications approach to database design. Dynamic data fusion enables a plug-and-play approach to data and insights-exploration. Datasets are kept logically separated and dynamically fused together at the point of query. This major step forward in database design eliminates the need for multiple copies of data, dramatically reduces storage costs, and enables an agile approach to data-powered product development.
CPU speed and memory size have increased rapidly over time. But data throughput between them remains low, creating a communication bottleneck which increases with every new generation of CPU.
GeoSpock overcomes the bottleneck with an embarrassingly parallel architecture simultaneously employing many disks and processing cores.
Best of all, our commodity hardware approach avoids expensive niche hardware and keeps costs down.