No single point of failure foryour public or private cloud
StorPool provides unmatched levels of high availability. This is due to its fully distributed shared-nothing architecture and remarkable design.
Server / rack level redundancy
Unlike traditional storage systems which have single points of failure, or dual redundancy components, StorPool takes redundancy and availability to a whole new level. Where others think in terms of components, StorPool thinks of complete servers or entire racks. Trying to make a single component (an SSD for example), as reliable as possible is a losing game, as the costs grows exponentially, whilst the reliability increase diminishes. This is why StorPool uses cost effective, standard, off the shelf components and expects and can tolerate failures. Losing a drive, server or sometimes an entire rack is something that StorPool’s distributed architecture handles by design.
High availability through intelligence in the data layer
As explained earlier StorPool uses a replication mechanism that slices and stores copies of the data on different servers or racks. For primary high performance storage, this solution has many advantages, compared to other alternatives. Most storage systems cannot tolerate the loss of an entire storage array, or a large number of the storage nodes in a cluster. StorPool on the other hand provides considerably higher levels of high availability, because it can easily survive a node, or in some cases, even a third of the storage cluster to go down. In case of drive, server or other component failure, StorPool uses another copy of the data, located on another server, (or rack) and none of the data is ever lost or temporarily unavailable.
- cost savings
- decreased efforts of your support team
- increased customer satisfaction
- no downtime during maintenance
- better performance
- better reliability