Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Proceedings - International Symposium on Computer Architecture/Proceedings

SATIC: An Optimizing Ising Compiler for SAT(isfiability)

Ahmet Efe, Hüsrev Cılasun, Abhimanyu Kumar, Nafisa Sadaf Prova, Ziqing Zeng, Tahmida Islam, R.S. Yin, Chaohui Li, Peter Kreye, Chris H. Kim, Sachin S. Sapatnekar, Ulya R. Karpuzcu

Published Jun 1, 2026• 1 citation
Publisher

Abstract

Ising machines show great potential as hardware solvers for combinatorial optimization problems such as Boolean Satisfiability (SAT) - one of the hardest classic problems of practical importance. Unlocking this potential, however, is only possible by bridging the gap between SAT problem characteristics and Ising machine specifics. In this paper, we introduce a novel optimizing compiler featuring a bag of powerful heuristic tricks, SATIC, toward this goal. We evaluate SATIC on a representative manufactured Ising chip. Our measurements show that SATIC enables Ising machines to solve up to$\text{73} \times$larger SAT problems than the hardware capacity would otherwise allow, demonstrating unmatched performance and scalability.

DOI
10.1109/isca66397.2026.00165

Related papers

Augmenting an electronic Ising machine to effectively solve boolean satisfiability

Anshujit Sharma, Matthew Burns, Andrew Hahn +1 more

With the slowdown of improvement in conventional von Neumann systems, increasing attention is paid to novel paradigms such as Ising machines. They have very different approach to solving combinatorial optimization problems. Ising machines have shown great potential in solving binary optimization problems like MaxCut. In this paper, we present an analysis of these systems in boolean satisfiability (SAT) problems. We demonstrate that, in the case of 3-SAT, a basic architecture fails to produce meaningful acceleration, largely due to the relentless progress made in conventional SAT solvers. Nevertheless, careful analysis attributes part of the failure to the lack of two important components: cubic interactions and efficient randomization heuristics. To overcome these limitations, we add proper architectural support for cubic interaction on a state-of-the-art Ising machine. More importantly, we propose a novel semantic-aware annealing schedule that makes the search-space navigation much more efficient than existing annealing heuristics. Using numerical simulations, we show that such an "Augmented" Ising Machine for SAT is projected to outperform state-of-the-art software-based, GPU-based and conventional hardware SAT solvers by orders of magnitude.

arXiv:2305.01623•Scientific Reports•Dec 21, 2023•27 citations

On the Limits of Ising Machines for Circuit-Derived SAT

Ahmet Efe, Abhimanyu Kumar, Nafisa Sadaf Prova +9 more

Emerging Ising machines are promising for solving computationally hard optimization problems, yet their limits on structured, circuit-derived satisfiability (SAT) problems remain poorly understood. Using semiprime factorization as a representative benchmark, we show that tight constraints, when mapped into optimization form, fundamentally distort the energy landscape, and that these distortions are amplified when problems are decomposed to fit limited Ising machine capacity. To address this, we propose a hybrid flow that offloads Ising-harmful structure to lightweight preprocessing while reserving the genuinely hard search for the Ising machine. We further show that generic, circuit-structure-unaware decomposition is insufficient for circuit-derived instances, and that structure-aware partitioning is essential. These findings identify constraint handling as a central obstacle, highlighting hybrid hardware-software approaches as the path forward for scaling Ising machines to real-world SAT workloads. Evaluated on fabricated 45-spin all-to-all Ising chips, our flow extends solvable problem sizes from 8-bit (94 variables) to 11-bit (190 variables) without any hardware changes.

arXiv:2511.21046•Nov 26, 2025•1 citation

Boolean satisfiability verification by quantum annealing

Maria Maltseva, Александр Сергеевич Румянцев, Mariya Maltseva +1 more

The Boolean satisfiability problem (SAT) is widely applied in modern tasks of microelectronics design automation. An important application of such a task is the design of logic circuits with specified characteristics from a given set of logic gates. To solve an SAT problem, both standard solvers and modern optimization tools can be used, including a relatively new hardware tool for solving the quantum annealing problem. The solution of this problem can be performed on a quantum annealing system by constructing a socalled Boolean network according to special rules and selecting an appropriate objective function. The article discusses the practical application of quantum annealing to the construction of short logic circuits by solving an SAT problem based on the methods proposed in the literature, and also analyzes the practical aspects of applying such methods to the construction of quantum circuits.

Proceedings of the Karelian Research Centre of the Russian Academy of Sciences•Jun 23, 2023•1 citation

An Experimental Evaluation of Ising Solvers for SAT problems

Tomohiro Sonobe

The development of Ising solvers, which search for solutions to Ising models and Quadratic Unconstrained Binary Optimization (QUBO), has spurred research into their application across various problems. In this paper, we present a computational experiment analyzing the performance of multiple Ising solvers on the satisfiability problem (SAT) and its derivative problems, MAX2SAT and NAE3SAT. While the SAT problem was found to be difficult to solve exactly, the performance of the Ising solvers on MAX2SAT and NAE3SAT was comparable to or better than that of existing dedicated solvers.

•Jul 24, 2025•0 citations

HETRI: Heterogeneous Ising Multiprocessing

Ahmet Efe, Hüsrev Cılasun, Abhimanyu Kumar +4 more

Ising machines are effective solvers for complex combinatorial optimization problems. The idea is mapping the optimal solution(s) to a combinatorial optimization problem to the minimum energy state(s) of a physical system, which naturally converges to a minimum energy state upon perturbance. The underlying mathematical abstraction, the Ising model, can capture the dynamic behavior of different physical systems by mapping each problem variable to a spin which can interact with other spins. Ising model as a mathematical abstraction can be mapped to hardware using traditional devices. In this paper we instead focus on Ising machines which represent a network of physical spins directly implemented in hardware using, e.g., quantum bits or electronic oscillators. To eliminate the scalability bottleneck due to the mismatch in problem vs. Ising machine size and connectivity, in this paper we make the case for HETRI: Heterogeneous Ising Multiprocessing. HETRI organizes the maximum number of physical spins that the underlying technology supports in Ising cores; and multiple independent Ising cores, in Ising chips. Ising cores in a chip feature different inter-spin connectivity or spin counts to match the problem characteristics. We provide a detailed design space exploration and quantify the performance in terms of time or energy to solution and solution accuracy with respect to homogeneous alternatives under the very same hardware budget and considering the very same spin technology.

arXiv:2410.23517•Oct 30, 2024•0 citations

Tripartite Kerr Soliton Ising Machine for Combinatorial Optimization

Nitesh Chauhan, Yan Jin, Jizhao Zang +1 more

We introduce and explore a Kerr soliton Ising machine with all-to-all connectivity and up to tripartite (cubic) interactions, unlocking efficient approaches for combinatorial optimization. The machine is an ensemble of solitons in a Kerr resonator, where a programmable optoelectronic feedback circuit generates amplitude mixing terms involving up to three arbitrary soliton spins in the system Hamiltonian. This capability offers a key advantage for solving canonical NP-complete problems like Boolean satisfiability (SAT) with three literal clauses, substantially reducing the spin overhead compared to quadratic-only Ising mappings. To explore our cubic Ising machine, we solve randomly generated 3-SAT instances of 250 variables and 1000 clauses, approaching the intrinsic satisfiability threshold. Our results highlight Kerr solitons as a flexible and energy-efficient analog computing substrate, with the potential to address challenges in combinatorial optimization, artificial intelligence, and machine learning.

Physical Review Letters•May 18, 2026•0 citations
Ising machines show great potential as hardware solvers for combinatorial optimization problems such as Boolean Satisfiability (SAT) - one of the hardest classic problems of practical importance. Unlocking this potential, however, is only possible by bridging the gap between SAT problem characteristics and Ising machine specifics. In this paper, we introduce a novel optimizing compiler featuring a bag of powerful heuristic tricks, SATIC, toward this goal. We evaluate SATIC on a representative manufactured Ising chip. Our measurements show that SATIC enables Ising machines to solve up to73×\text{73} \times73×larger SAT problems than the hardware capacity would otherwise allow, demonstrating unmatched performance and scalability.
With the slowdown of improvement in conventional von Neumann systems, increasing attention is paid to novel paradigms such as Ising machines. They have very different approach to solving combinatorial optimization problems. Ising machines have shown great potential in solving binary optimization problems like MaxCut. In this paper, we present an analysis of these systems in boolean satisfiability (SAT) problems. We demonstrate that, in the case of 3-SAT, a basic architecture fails to produce meaningful acceleration, largely due to the relentless progress made in conventional SAT solvers. Nevertheless, careful analysis attributes part of the failure to the lack of two important components: cubic interactions and efficient randomization heuristics. To overcome these limitations, we add proper architectural support for cubic interaction on a state-of-the-art Ising machine. More importantly, we propose a novel semantic-aware annealing schedule that makes the search-space navigation much more efficient than existing annealing heuristics. Using numerical simulations, we show that such an "Augmented" Ising Machine for SAT is projected to outperform state-of-the-art software-based, GPU-based and conventional hardware SAT solvers by orders of magnitude.
Emerging Ising machines are promising for solving computationally hard optimization problems, yet their limits on structured, circuit-derived satisfiability (SAT) problems remain poorly understood. Using semiprime factorization as a representative benchmark, we show that tight constraints, when mapped into optimization form, fundamentally distort the energy landscape, and that these distortions are amplified when problems are decomposed to fit limited Ising machine capacity. To address this, we propose a hybrid flow that offloads Ising-harmful structure to lightweight preprocessing while reserving the genuinely hard search for the Ising machine. We further show that generic, circuit-structure-unaware decomposition is insufficient for circuit-derived instances, and that structure-aware partitioning is essential. These findings identify constraint handling as a central obstacle, highlighting hybrid hardware-software approaches as the path forward for scaling Ising machines to real-world SAT workloads. Evaluated on fabricated 45-spin all-to-all Ising chips, our flow extends solvable problem sizes from 8-bit (94 variables) to 11-bit (190 variables) without any hardware changes.
The Boolean satisfiability problem (SAT) is widely applied in modern tasks of microelectronics design automation. An important application of such a task is the design of logic circuits with specified characteristics from a given set of logic gates. To solve an SAT problem, both standard solvers and modern optimization tools can be used, including a relatively new hardware tool for solving the quantum annealing problem. The solution of this problem can be performed on a quantum annealing system by constructing a socalled Boolean network according to special rules and selecting an appropriate objective function. The article discusses the practical application of quantum annealing to the construction of short logic circuits by solving an SAT problem based on the methods proposed in the literature, and also analyzes the practical aspects of applying such methods to the construction of quantum circuits.
The development of Ising solvers, which search for solutions to Ising models and Quadratic Unconstrained Binary Optimization (QUBO), has spurred research into their application across various problems. In this paper, we present a computational experiment analyzing the performance of multiple Ising solvers on the satisfiability problem (SAT) and its derivative problems, MAX2SAT and NAE3SAT. While the SAT problem was found to be difficult to solve exactly, the performance of the Ising solvers on MAX2SAT and NAE3SAT was comparable to or better than that of existing dedicated solvers.
Ising machines are effective solvers for complex combinatorial optimization problems. The idea is mapping the optimal solution(s) to a combinatorial optimization problem to the minimum energy state(s) of a physical system, which naturally converges to a minimum energy state upon perturbance. The underlying mathematical abstraction, the Ising model, can capture the dynamic behavior of different physical systems by mapping each problem variable to a spin which can interact with other spins. Ising model as a mathematical abstraction can be mapped to hardware using traditional devices. In this paper we instead focus on Ising machines which represent a network of physical spins directly implemented in hardware using, e.g., quantum bits or electronic oscillators. To eliminate the scalability bottleneck due to the mismatch in problem vs. Ising machine size and connectivity, in this paper we make the case for HETRI: Heterogeneous Ising Multiprocessing. HETRI organizes the maximum number of physical spins that the underlying technology supports in Ising cores; and multiple independent Ising cores, in Ising chips. Ising cores in a chip feature different inter-spin connectivity or spin counts to match the problem characteristics. We provide a detailed design space exploration and quantify the performance in terms of time or energy to solution and solution accuracy with respect to homogeneous alternatives under the very same hardware budget and considering the very same spin technology.
We introduce and explore a Kerr soliton Ising machine with all-to-all connectivity and up to tripartite (cubic) interactions, unlocking efficient approaches for combinatorial optimization. The machine is an ensemble of solitons in a Kerr resonator, where a programmable optoelectronic feedback circuit generates amplitude mixing terms involving up to three arbitrary soliton spins in the system Hamiltonian. This capability offers a key advantage for solving canonical NP-complete problems like Boolean satisfiability (SAT) with three literal clauses, substantially reducing the spin overhead compared to quadratic-only Ising mappings. To explore our cubic Ising machine, we solve randomly generated 3-SAT instances of 250 variables and 1000 clauses, approaching the intrinsic satisfiability threshold. Our results highlight Kerr solitons as a flexible and energy-efficient analog computing substrate, with the potential to address challenges in combinatorial optimization, artificial intelligence, and machine learning.