EXACluster OS forms the platform for EXASolution. EXACluster OS is based on the standard Linux kernel. EXASOL extended this kernel by adding cluster-specific functionality.

EXACluster OS guarantees high availability of the overall system and offers cluster-wide efficient storage mechanisms. EXACluster OS provides additional services, such as centralized cluster administration and monitoring.

The nodes in the cluster are peers and can assume any task.

High availability

If a cluster node fails, a standby node takes over its function. EXACluster OS thus ensures that operation continues uninterrupted. It only takes a few seconds to resume operation. If a node fails, the data of the failed node are simply transferred to the standby node and the user can continue to work normally.

The number of standby nodes in the system depends on the customer’s availability requirements.

 

Efficient storage

EXASolution distributes the data uniformly over all nodes in the cluster (shared nothing). In order to guarantee high availability in the event of a node failure, EXACluster OS ensures that the data are stored redundantly. Data are stored on local hard disks, resulting in very high cluster-wide throughput for read and write operations.

Thanks to this storage method and in-memory data processing, expensive storage systems such as SAN are no longer required.

 

Standard server

The EXASolution system operates on economical standard hardware. A cluster of powerful Intel servers mounted in 19-inch racks is usually used.

 A typical configuration of such a cluster node consists of

 

• 2 Intel Xeon processors, each with two or four cores
• 8 to 64 GB RAM
• 4 to 12 SATA hard disks

The cluster is connected via GBit Ethernet.