Efficient Computation of Hashes

The sequential computation of hashes at the core of many distributed storage systems and found, for example, in grid services can hinder efficiency in service quality and even pose security challenges that can only be addressed by the use of parallel hash tree modes.

The main contributions of this paper are, first, the identification of several efficiency and security challenges posed by the use of sequential hash computation based on the Merkle-Damgard engine. In addition, alternatives for the parallel computation of hash trees are discussed, and a prototype for a new parallel implementation of the Keccak function, the SHA-3 winner, is introduced.