cluster

A set of machines located in one place and acting as a single "virtual machine" even though they are not as tightly integrated as a single SMP machine. Clusters often involve sharing disks but not memory, and are as often used to implement high availability as for performance reasons.

For something that's _not_ a cluster, but rather an example of shameless self-promoting abuse of the term to sound cool, see Debian's CLOWN project.

Book by Piers Anthony.

Deals with our green-skinned hero, Flint of Outworld, attempting to thwart the threat of invasion of our galaxy by Galaxy Andromeda...his high Kirlian aura allows him to possess other people's bodies for months at a time with the proper equipment. So, he acts as an ambassador to other races, trying to get them to band together and stop the Andromedan invasion.

Much less hokey then it sounds--one of the greatest books of all-time. Later expanded into a series which followed various decendants of Flint on their journeys.

A cluster is a set of computers (usually called nodes) that act together to create a Single System Image. The image can be from a lot of different viewpoints - service availability (the service running on the cluster is available even if a node is down), performance (you can submit a job to the cluster, and it can run as well on any node), administration (the administrator don't have to care about the difference between indivdual nodes), or just about any other logical point you can view a computing system from.

The most common reason for creating a cluster instead of getting a really fast machine is High Availability. Due to the increased cost of top-of-the-line computers compared to the low end, you can usually get the same performance from a cluster of smaller machines for about the same cost - and get High Availability almost for free.

A statistical technique for figuring out how to clump things together. This seems like an easy problem which you could just solve visually, but sometimes the clusters are not obvious, especially if the elements you want to cluster are scattered over a large area.

There are a number of different variations on this technique, but they all essentially work the same way. First, find the two elements which are closest together, and make a new, single element out of them. That's one cluster. Now, repeat. Repeat again until everything's assigned to a cluster, or you have the right number of clusters (decided a priori), or whatever. The variation comes in deciding the position of the new cluster you create. You can make it the position of the first point you chose, for instance, or the average position of all the original points in the cluster, or the average position of the new point and the current position, and so forth. These subtle differences can lead, in some cases, to dramtically different sets of clusters.

When speaking of disk space, a cluster is the smallest possible amount of disk space the file system can allocate.

A file always takes one or more clusters of disk space. For example, if you store a file of 600 bytes to filesystem that allocates disk space in 1024-byte clusters, the file will physically take 1024 bytes. The remaining 424 bytes cannot be used. A file of 1092 bytes will take 2048 bytes physically, and so on.

The reason behind this? Efficiency, of course.

The clusters are mostly an issue in FAT. MS-DOS used rather silly (but efficient at the day) way of organizing the filesystem to clusters whose size depended on disk size. Floppies used 512-byte clusters and hard disks used clusters of multiplies of 1024 - biggest drives needed clusters of 32 kilobytes (or even more, I can't remember)! The amount of disk space wasted by FAT16 was rather tremendous.

FAT32, introduced in Windows 95 OSR2, helped the situation somewhat by decreasing the cluster size on hard drives to 4096 bytes.

Other OSes and filesystems don't need to much worry about clusters and lost space...

See also slack.

Clus"ter (?), n. [AS. cluster, clyster; cf. LG. kluster (also Sw. & Dan. klase a cluster of grapes, D. klissen to be entangled?.)]

1.

A number of things of the same kind growing together; a bunch.

Her deeds were like great clusters of ripe grapes, Which load the bunches of the fruitful vine. Spenser.

2.

A number of similar things collected together or lying contiguous; a group; as, a cluster of islands.

"Cluster of provinces."

Motley.

3.

A number of individuals grouped together or collected in one place; a crowd; a mob.

As bees . . . Pour forth their populous youth about the hive In clusters. Milton.

We loved him; but, like beasts And cowardly nobles, gave way unto your clusters, Who did hoot him out o' the city. Shak.

 

© Webster 1913.

---

Clus"ter, v. i. [imp. & p. p. Clustered (?); p. pr. & vb. n. Clustering.]

To grow in clusters or assemble in groups; to gather or unite in a cluster or clusters.

His sunny hair Cluster'd about his temples, like a god's. Tennyson.

The princes of the country clustering together. Foxe.

 

© Webster 1913.

---

Clus"ter, v. t.

To collect into a cluster or clusters; to gather into a bunch or close body.

Not less the bee would range her cells, . . . The foxglove cluster dappled bells. Tennyson.

Or from the forest falls the clustered snow. Thomson.

Clustered column Arch., a column which is composed, or appears to be composed, of several columns collected together.

 

© Webster 1913.