A path toward ultra-low-energy computing

DeBenedictis, Erik P.; Frank, Michael P.; Ganesh, Natesh; Anderson, Neal G.

doi:10.1109/icrc.2016.7738677

Cited by 20 publications

(19 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The key insight that allows us to advance from the traditional theory to the more general one is simply the realization, discussed in [13], that a computation per se consists of not just a choice of an abstract computational operation to be carried out, but also a specific statistical operating context in which that operation is to be applied. Without considering the specific statistical context, one cannot calculate the initial and final computational entropies with any degree of accuracy, and therefore, one cannot correctly infer whether any entropy is in fact ejected from the computational state in the course of the computation.…”

Section: Reformulating Reversible Computing Theorymentioning

confidence: 99%

“…A numerical example illustrating how the ∆S nc calculation comes out in a specific case where the probability of violating the precondition for reversibility is small can be found in [13], which also discussed the more general issue that the initial-state probabilities must be taken into account in order to properly apply Landauer's Principle.…”

Section: Fundamental Theorem Of Generalized Reversible Computingmentioning

confidence: 99%

See 1 more Smart Citation

Foundations of Generalized Reversible Computing

Frank

2017

Reversible Computation

Self Cite

View full text Add to dashboard Cite

Landauer's Principle that the loss of information from a computation corresponds to an increase in entropy can be expressed as a rigorous theorem of mathematical physics. However, carefully examining its detailed formulation reveals that the traditional definition identifying logically reversible computational operations with bijective transformations of the full digital state space is actually not the most general characterization, at the logical level, of the complete set of classical computational operations that can be carried out physically with asymptotically zero energy dissipation. To derive the correct set of necessary logical conditions for physical reversibility, we must take into account the effect of initial-state probabilities when applying the detailed form of the Principle. The minimal logical-level requirement for the physical reversibility of deterministic computational operations turns out to be that only the subset of initial states that are assigned nonzero probability in a given statistical operating context must be transformed one-to-one into final states. Consequently, any computational operation can be seen as conditionally reversible, relative to any sufficiently-restrictive precondition on its initial state, and the minimum average dissipation required for any deterministic operation by Landauer's Principle asymptotically approaches zero in contexts where the probability of meeting any preselected one of its suitable preconditions approaches unity. The concept of conditional reversibility facilitates much simpler designs for asymptotically thermodynamically reversible computational devices and circuits, compared to designs that are restricted to using only fully-bijective operations such as Fredkin/Toffoli type operations. Thus, this more general framework for reversible computing provides a more effective theoretical foundation to use for the design of practical reversible computing hardware than does the more restrictive traditional model of reversible logic. In this paper, we formally develop the theoretical foundations of the generalized model, and briefly survey some of its applications.

show abstract

Section: Reformulating Reversible Computing Theorymentioning

confidence: 99%

Section: Fundamental Theorem Of Generalized Reversible Computingmentioning

confidence: 99%

Foundations of Generalized Reversible Computing

Frank

2017

Reversible Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…Landaurer's limit is not yet a serious consideration for modern classical computers because current technology already wastes several orders of magnitude more energy, due to conventional engineering imperfections, than is required by Landaurer's limit. But, as part of the progress associated with Moore's law, the amount of energy wasted has been decreasing exponentially each year, and eventually Landaurer's limit will become important for continuing technological progress in classical computing (DeBenedictis et al, 2016(DeBenedictis et al, , 2017. Once that happens, classical computers will likely need to become reversible (or partially reversible) to continue improving.…”

Section: Reversible Computingmentioning

confidence: 99%

Forecasting timelines of quantum computing

Sevilla,

Riedel

2020

Preprint

View full text Add to dashboard Cite

We consider how to forecast progress in the domain of quantum computing. For this purpose we collect a dataset of quantum computer systems to date, scored on their physical qubits and gate error rate, and we define an index combining both metrics, the generalized logical qubit.We study the relationship between physical qubits and gate error rate, and tentatively conclude that they are positively correlated (albeit with some room for doubt), indicating a frontier of development that trades-off between them.We also apply a log-linear regression on the metrics to provide a tentative upper bound on how much progress can be expected over time. Within the (generally optimistic) assumptions of our model, including the key assumption that exponential progress in qubit count and gate fidelity will continue, we estimate that that proof-of-concept fault-tolerant computation based on superconductor technology is unlikely (<5% probability) to be exhibited before 2026, and that quantum devices capable of factoring RSA-2048 are unlikely (<5% probability) to exist before 2039. It is of course possible that these milestones will in fact be reached earlier, but that this would require faster progress than has yet been seen.

show abstract

“…Entropy has been applied to VLSI (very large scale integration) to handle complexity of signals and circuits, see [37,51], for everything from reliable memory [52] to reversible logic [38,28] to decoupling capacitor placement [56]. Entropy is also not surprisingly used in a number of energy and power related ways, such as partitioning for power estimation [14], VLSI floorplanning for temperature reduction [8], and low-power computing [29,11]. Link entropy is a standard measure of evaluating network connections, see [55].…”

Section: Entropy In Computationmentioning

confidence: 99%

Entropy of Parallel Execution and Communication

Gomez

Cai

Schubert

2017

IJNC

View full text Add to dashboard Cite

We propose a definition of parallel state, derive a phase space from this state, and calculate the entropy of states and full executions using combinatorial analysis. A main contribution of this work is the introduction of an experimentally measurable phase space, which we then use to analyze execution states, ensembles of states, and ensembles of complete executions. We show that the entropy analysis reveals both expected and unexpected features of execution, and application of principal component analysis shows capability to extract execution details at the level of individual process states, as well as reveal hardware properties such as network or memory communications.

show abstract

A path toward ultra-low-energy computing

Cited by 20 publications

References 11 publications

Foundations of Generalized Reversible Computing

Foundations of Generalized Reversible Computing

Forecasting timelines of quantum computing

Entropy of Parallel Execution and Communication

Contact Info

Product

Resources

About