Insightful Business Intelligence

Engage - Data Warehousing

Ryder System's DaaS (data warehousing as a service) utilizes a shared-nothing MPP (massively parallel processing) architecture that has been built from the ground for BI.

Data is automatically partitioned across multiple servers with each server managing part of the overall data. The servers are able to process queries in a fully parallel manner enabling massive scalability and near linear query performance for multi terabyte data sets. The degree of parallelism and overall scalability that this allows for far exceeds general purpose database systems.

The system is architectured to have no single point of failure via the use of multiple levels of redundancy and integrity checking. At the lowest level, RAID-1+0 or RAID-5 storage is used to detect and recover from disk failures. At the system level, data is replicated across servers to ensure that the loss of a machine will not impact overall system availability. At the applications level, hot standby servers are used for automated failover.

We use Redhat Enterprise Linux and commodity hardware from Dell.