Non-volatile Position Verification for Multi-node Networks
Original Publication Date: 2000-Nov-01
Included in the Prior Art Database: 2003-Jun-19
When a multinode network is configured the position of the nodes is often important. Consider the case of a storage sub-system with raid arrays composed of multiple storage nodes. These array components may need to be located across more than one physical location in order to ensure continued access to data in the event of a catastrophic failure in one of the locations. e.g. a split site raid 10 configuration where a raid 0 array in one location is mirrored as a raid 0 array in another location. In such a configuration it can be important that the components of each array are physically located in the correct location. Consider a network consisting of one initiator and 6 storage nodes. Storage nodes 1 to 3 are configured as a raid 0 array at site A and storage nodes 4 to 6 are configured as a raid 0 array at site B. The raid 0 array at site B is a mirrored copy of the raid 0 array at site A. i.e. the 6 storage nodes constitute a single raid 10 array split across two sites. If site A or site B suffers a catastrophic failure the data is still available at the other site. However if an error was made during maintenance of the network such that for example storage node 3 was located at site B and storage node 6 was located at site A and one of the sites suffers a catastrophic failure the data is completely lost. Such an error would be easy to make during maintenance and would not be evident at the time. In the case of a multi-initiator network composed of many storage nodes spread over several sites with hot spares configured to take over in the event of individual storage node failure it can quickly become difficult to ensure routing integrity. This invention solves this problem by storing the position of each storage device in a multinode network on that storage device.