The Cost of Energy-Efficiency in Digital Hardware: The Trade-Off between Energy Dissipation, Energy–Delay Product and Reliability in Electronic, Magnetic and Optical Binary Switches

Abstract

Binary switches, which are the primitive units of all digital computing and information processing hardware, are usually benchmarked on the basis of their ‘energy–delay product’, which is the product of the energy dissipated in completing the switching action and the time it takes to complete that action. The lower the energy–delay product, the better the switch (supposedly). This approach ignores the fact that lower energy dissipation and faster switching usually come at the cost of poorer reliability (i.e., a higher switching error rate) and hence the energy–delay product alone cannot be a good metric for benchmarking switches. Here, we show the trade-off between energy dissipation, energy–delay product and error–probability for an electronic switch (a metal oxide semiconductor field effect transistor), a magnetic switch (a magnetic tunnel junction switched with spin transfer torque) and an optical switch (bistable non-linear mirror). As expected, reducing energy dissipation and/or energy–delay product generally results in increased switching error probability and reduced reliability.

Applications of nanomagnets as dynamical systems: I

“…Unfortunately, it turns out that magnetic devices are either too energyinefficient or too error prone for Boolean logic. The energy dissipated to switch a modern-day transistor (a volatile switch) is about 100 aJ, which could probably be lowered to about 10 aJ [42,[44][45][46] and the switching error rate is about 10 -15 [47], while the energy dissipated to switch an MTJ (a non-volatile switch) with STT is at least 10 fJ with a Electrons tunnel between the two ferromagnetic layers though the spacer and because of spin-dependent tunneling, the resistance of the device measured between the ferromagnetic layers is low when the latter have parallel magnetizations and high when they have antiparallel magnetizations. These two resistance states encode the binary bits 0 and 1.…”

“…There seems to be always an unavoidable trade-off between energy-efficiency and reliability when it comes to binary switches. One can only be purchased at the cost of the other and this is true of electronic, magnetic and even optical switches [50]. Boolean logic has stringent requirements for reliability.…”

“…On the other hand, voltage (or electric-field) induced magnetization switching is much more energy-efficient, albeit also less reliable (higher switching error probability). There is always a trade-off between energy cost and error resilience, which is also true of electronic devices like the transistor [44,45].…”

Quantum Phase Transition in the Spin Transport Properties of Ferromagnetic Metal-Insulator-Metal Hybrid Materials

“…Although several concerted efforts have been focused on the study of different 2D materials as the tunnel barrier layer in pMTJs [ 12 , 13 , 14 ], neither the nature of the underlying magnetoelectric coupling mechanisms nor the roles of the tunnel barrier layer in the transmitted spin current are well-understood. Gaining insights into such phenomena is a crucial first step toward achieving a lower energy-delay product performance [ 15 ].…”

“…These include the spontaneous breaking of the chiral symmetry of carrier spin fields in the presence of electromagnetic fields [ 17 ], non-conservation of the rotational symmetry of the spin degree of freedom for carriers at constituent heterobilayer interfaces relative to the spin quantization axis, and the non-conservation of both parity and helicity when spin is projected upon the angular momentum. These broken symmetries are sources of spin anisotropy when the pMTJ design integrates hard ferromagnetic leads with an insulating tunnel barrier layer [ 15 , 18 ], without incorporating a topological insulator to provide the required spin-momentum locking [ 19 , 20 ]. Although insights into the spin field are gained by analyzing the Fermi surface topology for spin carriers [ 21 ], field-dependence of the tunneling magnetoresistance (TMR) and its anisotropy [ 22 ], and sensitivity of the interfacial spin backflow to tunnel barrier material [ 23 ], strategies for mitigating the challenge of energy-delay performance in spintronic devices are still not yet clear.…”