TY - CHAP
T1 - Proposal for a Quantum-Based Memory for Storing Classical Information and the Connection Between Molecular Dynamics Simulations and the Landauer’s Principle
AU - Batle, Josep
AU - Elhoseny, Mohamed
AU - Farouk, Ahmed
N1 - Publisher Copyright:
© 2018, Springer International Publishing AG.
PY - 2018
Y1 - 2018
N2 - The development of high-capacity memory devices plays an increasingly important role in modern society. High capacities in information storage constitutes a key resource for dealing with the everyday generation of information, as well as for handling the so called Big Data generated in different scientific and technological scenarios. By combining precision metrology and quantum devices such as quantum dots and quantum wires, we propose a quantum memory whose capacity depends on the particular architecture chosen, namely, linear or planar. We show that the geometric disposition of minimal quantum cells or chips is critical in having similar or dramatically outperformed information capacities as compared to current devices. This information is stored in the form of classical bits, though. Realization of such a quantum memory may solve a two-fold problem at the same time: unprecedented higher information capacity with undefined longevity. We shall obtain as well, by rigorously applying the definition of the exponentiation of a Hermitian matrix, the set of Hamiltonians whose evolution corresponds to the set of universal gates. Also, Landauer’s principle is a fundamental link between thermodynamics and information theory, which implies that the erasure of information comes at an energetic price, either in classical or quantum computation. In the present contribution we analyze to what extend the usual molecular dynamics (MD) simulation formalism can handle the Landauer’s bound $$k:BT\ln 2$$ in the simplest case of one particle treated classically. The erasure of one bit of information is performed by adiabatically varying the shape of a bistable potential in a full cycle. We will highlight the inadequacy of either the microcanonical or canonical ensemble treatments currently employed in MD simulations and propose potential solutions.
AB - The development of high-capacity memory devices plays an increasingly important role in modern society. High capacities in information storage constitutes a key resource for dealing with the everyday generation of information, as well as for handling the so called Big Data generated in different scientific and technological scenarios. By combining precision metrology and quantum devices such as quantum dots and quantum wires, we propose a quantum memory whose capacity depends on the particular architecture chosen, namely, linear or planar. We show that the geometric disposition of minimal quantum cells or chips is critical in having similar or dramatically outperformed information capacities as compared to current devices. This information is stored in the form of classical bits, though. Realization of such a quantum memory may solve a two-fold problem at the same time: unprecedented higher information capacity with undefined longevity. We shall obtain as well, by rigorously applying the definition of the exponentiation of a Hermitian matrix, the set of Hamiltonians whose evolution corresponds to the set of universal gates. Also, Landauer’s principle is a fundamental link between thermodynamics and information theory, which implies that the erasure of information comes at an energetic price, either in classical or quantum computation. In the present contribution we analyze to what extend the usual molecular dynamics (MD) simulation formalism can handle the Landauer’s bound $$k:BT\ln 2$$ in the simplest case of one particle treated classically. The erasure of one bit of information is performed by adiabatically varying the shape of a bistable potential in a full cycle. We will highlight the inadequacy of either the microcanonical or canonical ensemble treatments currently employed in MD simulations and propose potential solutions.
UR - http://www.scopus.com/inward/record.url?scp=85132890236&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-63639-9_13
DO - 10.1007/978-3-319-63639-9_13
M3 - Chapter
AN - SCOPUS:85132890236
T3 - Studies in Big Data
SP - 291
EP - 316
BT - Studies in Big Data
PB - Springer Science and Business Media Deutschland GmbH
ER -