An Efficient Data Integration Framework in Cloud Using MapReduce - Computational Intelligence Techniques for Comparative Genomics - page 126

rst,
data_octate is input data split of an octate size, and prime stands for selected FNV
prime.
There is FNV-1 also available; only difference between it and FNV-1a hash is
that the order of the multiply and xor will be altered, and prime and offset used by
both of them are same.

Here, H stands for hash, offset is initial vector that has to take at

3.3 Internal Operation

At a data node, if the user is authorized to access the cloud, he can make a data
transfer, where the data can be spitted into parts and for each, a message digest will
be generated which is called child digest; all child digests will be added to master
digest

le and are encrypted together with the original data of the owner and sent
back to him when he requests, which is shown in Fig. 3 .
Since only the latest checkpoint and operation log related are only needed, the
older ones can be deleted. The operation log and checkpoints are replicated to
provide for reliability by Secondary name node. The ECMF signi

cantly improves

its robustness.

Fig. 3 Overall architecture of hashing-based cloud environment