For enterprise data storage, clustering usually requires a shared storage system, such as a storage-area network (SAN). A mechanism to duplicate the data in the storage system that's directly attached to each node can also be used.
A distributed computing system consists of nodes that have processing power and memory to support that processing. A distributed system architecture is typically used to break a larger processing job into sub-processes that each node can handle simultaneously to shorten the job.
A mainframe system is a single-node compute system. Mainframe systems were originally designed with a single processing engine, but also have multiple processors that can operate somewhat independently.
To learn more about clustered storage:
- Take a crash course on clustered storage
- Learn how network-attached storage (NAS) can be made easier with clustered NAS
- Discover the differences between virtualization and clustering
BIO: Eric Slack is an analyst at Storage Switzerland
This was first published in May 2010