Cryptography is increasingly viewed as a critical technology to fulfill the requirements of security and authentication for information exchange between Internet applications. However, software implementations of cryptographic applications are unable to support the quality of service from a bandwidth perspective required by most Internet applications. As a result, various hardware implementations, from Application-Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), to programmable processors, were proposed to improve this inadequate quality of service. Although these implementations provide performances that are considered better than those produced by software implementations, they still fall short of addressing the bandwidth requirements of most cryptographic applications in the context of the Internet for two major reasons: (i) The majority of these architectures sacrifice flexibility for performance in order to reach the performance level needed for cryptographic applications. This lack of flexibility can be detrimental considering that cryptographic standards and algorithms are still evolving. (ii) These architectures do not consider the consequences of technology scaling in general, and particularly interconnect related problems. As a result, this thesis proposes an architecture that attempts to address the requirements of cryptographic applications by overcoming the obstacles described in (i) and (ii). To this end, we propose a new reconfigurable, two-dimensional, scalable architecture, called CRYPTARRAY, in which bus-based communication is replaced by distributed shared memory communication. At the physical level, the length of the wires will be kept to a minimum. CRYPTARRAY is organized as a chessboard in which the dark and light squares represent Processing Elements (PE) and memory blocks respectively. The granularity and resource composition of the PEs is specifically designed to support the computing operations encountered in cryptographic algorithms in general, and symmetric algorithms in particular. Communication can occur only between neighboring PEs through locally shared memory blocks. Because of the chessboard layout, the architecture can be reconfigured to allow computation to proceed as a pipelined wave in any direction. This organization offers a high computational density in terms of datapath resources and a large number of distributed storage resources that easily support a high degree of parallelism and pipelining. Experimental prototyping a small array on FPGA chips shows that this architecture can run at 80.9 MHz producing 26,968,716 outputs every second in static reconfiguration mode and 20,226,537 outputs every second in dynamic reconfiguration mode.