Sampling techniques are the means by which participants are selected to take part in any kind of research.
People in scientific research are not just selected at random (although "random sampling" is indeed one of the techniques!), the way they are chosen is an important variable of the research itself.
By nature, a sample is supposed to be a group which is representative of a larger population.
Cluster sampling occurs when "natural" groupings (known as clusters) are evident within a population, and a random sample is then taken from this cluster.
For example- if market research was to be conducted in an area of town with three streets, then the residents of each street would be the cluster. Then, if two families were selected from each cluster, that would be an example of cluster sampling.