Computing Reviews

Architectures of flexible symmetric key crypto engines--a survey:from hardware coprocessor to multi-crypto-processor system on chip
Bossuet L., Grand M., Gaspar L., Fischer V., Gogniat G. ACM Computing Surveys45(4):1-32,2013.Type:Article
Date Reviewed: 11/13/13

The authors present a survey of dedicated hardware introduced over the past decade for cryptographic operations, with a taxonomy of cryptographic hardware in terms of performance, flexibility, and security. The survey identified four main classes of cryptographic hardware: additions to general-purpose processors, cryptographic processors, cryptographic coprocessors, and cryptographic arrays. The paper provides essential information for anyone interested in the background and history of the field.

Much of the paper is focused on enumeration. The authors point to some interesting issues. They emphasize that it is necessary to protect secure information by enforcing information flow control between the main system and the cryptographic processing elements and data. This prevents leakage of key data during attacks and is the main point of consideration for the security dimension in their taxonomy. Another interesting point is the lack of a standard mechanism to compare performance across different hardware solutions for cryptography. This is a consequence of hardware heterogeneity rather than the result of a proliferation of cryptographic algorithms.

The sheer amount of detail that needs to be navigated by the reader makes this paper somewhat difficult to read, and parts of it may only be of interest to specialists in the area of cryptographic hardware. There is also a lack of treatment of hardware for public-key cryptosystems, and no treatment at all of accelerators for public-key ciphers such as RSA and elliptic curve cryptography (ECC). Furthermore, the paper provides very little discussion of accelerating signature generation. It is unclear whether this is because such systems are rare or because the authors consider the algorithmic operations to be subsumed by the private-key ciphers they do discuss, meaning a separate treatment is not required.

On the whole, however, this is a very useful paper for both beginners and specialists in the field.

Reviewer:  Amitabha Roy Review #: CR141725 (1401-0074)

Reproduction in whole or in part without permission is prohibited.   Copyright 2024 ComputingReviews.com™
Terms of Use
| Privacy Policy