An official website of the United States government
We report on a new class of Ising machines (IMs) that rely on coupled parametric frequency dividers (PFDs) as macroscopic artificial spins. Unlike the IM counterparts based on subharmonic-injection locking (SHIL), PFD IMs donot require strong injected continuous-wave signals or applied dc voltages. Therefore, they show a significantly lower power consumption per spin compared to SHIL-based IMs, making it feasible to accurately solve large-scale combinatorial optimization problems that are hard or even impossible to solve by using the current von Neumann computing architectures. Furthermore, using high quality factor resonators in the PFD design makes PFD IMs able to exhibit a nanowatt-level power per spin. Also, it remarkably allows a speedup of the phase synchronization among the PFDs, resulting in shorter time to solution and lower energy to solution despite the resonators’ longer relaxation time. As a proof of concept, a 4-node PFD IM has been demonstrated. This IM correctly solves a set of Max-Cut problems while consuming just 600 nanowatts per spin. This power consumption is 2 orders of magnitude lower than the power per spin of state-of-the-art SHIL-based IMs operating at the same frequency.
more »
« less
This content will become publicly available on August 19, 2026
Limitations in parallel Ising machine networks: Theory and practice
Analog Ising machines (IMs) occupy an increasingly prominent area of computer architecture research, offering high-quality, low-latency, and low-energy solutions to intractable computing tasks; however, IMs have a fixed capacity, with little to no utility in out-of-capacity problems. Previous works have proposed parallel, multi-IM architectures to circumvent this limitation [A. Sharma, , in , ISCA ’22 (Association for Computing Machinery, New York, NY, USA, 2022), p. 508; R. Santos, , Enhancing quantum annealing via entanglement distribution, ArXiv:2212.02465]. In this work, we theoretically and numerically investigate trade-offs in parallel IM networks to guide researchers in this burgeoning field. We propose formal models of parallel IM execution models, and we then provide theoretical guarantees for probabilistic convergence. Numerical experiments illustrate our findings and provide empirical insights into the high- and low-synchronization-frequency regimes. We also provide practical heuristics for parameter and model selection, informed by our theoretical and numerical findings.
more »
« less
- PAR ID:
- 10634771
- Publisher / Repository:
- APS
- Date Published:
- Journal Name:
- Physical Review Applied
- Volume:
- 24
- Issue:
- 2
- ISSN:
- 2331-7019
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
ABSTRACT Lipids, indispensable yet structurally intricate biomolecules, serve as critical regulators of cellular function and disease progression. Conventional lipidomics, constrained by limited resolution for isomeric and low‐abundance species, has been transformed by ion mobility‐mass spectrometry (IM‐MS). This technology augments analytical power through enhanced orthogonal separation, collision cross‐section (CCS)‐based identification, and improved sensitivity. This review examines the transformative advances in IM‐MS‐driven lipidomics, focusing on three major pillars: (1) a critical evaluation of leading ion mobility spectrometry (IMS) platforms, emphasizing innovative instrument geometries and breakthroughs in resolving lipid isomers; (2) an exploration of lipid CCS databases and predictive frameworks, spotlighting computational modeling and machine learning strategies that synergize experimental data with molecular representations for high‐confidence lipid annotation; (3) emerging multi‐dimensional lipidomics workflows integrating CCS with liquid chromatography‐MS/MS to boost identification and depth, alongside mass spectrometry imaging for spatially resolved lipidomics. By unifying cutting‐edge instrumentation, computational advances, and biological insights, this review outlines a roadmap for leveraging IM‐MS to unravel lipidome complexity, catalyzing biomarker discovery and precision medicine innovation.more » « less
-
Influence Maximization (IM) is a crucial problem in data science. The goal is to find a fixed-size set of highly influentialseedvertices on a network to maximize the influence spread along the edges. While IM is NP-hard on commonly used diffusion models, a greedy algorithm can achieve (1 - 1/e)-approximation by repeatedly selecting the vertex with the highestmarginal gainin influence as the seed. However, we observe two performance issues in the existing work that prevent them from scaling to today's large-scale graphs: space-inefficient memorization to estimate marginal gain, and time-inefficient seed selection process due to a lack of parallelism. This paper significantly improves the scalability of IM using two key techniques. The first is asketch-compressiontechnique for the independent cascading model on undirected graphs. It allows combining the simulation and sketching approaches to achieve a time-space tradeoff. The second technique includes new data structures for parallel seed selection. Using our new approaches, we implementedPaC-IM: Parallel and Compressed IM. We comparePaC-IMwith state-of-the-art parallel IM systems on a 96-core machine with 1.5TB memory.PaC-IMcan process the ClueWeb graph with 978M vertices and 75B edges in about 2 hours. On average, across all tested graphs, our uncompressed version is 5--18x faster and about 1.4x more space-efficient than existing parallel IM systems. Using compression further saves 3.8x space with only 70% overhead in time on average.more » « less
-
Low molecular weight metabolites are essential for defining the molecular phenotypes of cells. However, spatial metabolomics tools often lack the sensitivity, specify, and spatial resolution to provide comprehensive descriptions of these species in tissue. MALDI imaging mass spectrometry (IMS) of low molecular weight ions is particularly challenging as MALDI matrix clusters are often nominally isobaric with multiple metabolite ions, requiring high resolving power instrumentation or derivatization to circumvent this issue. An alternative to this is to perform ion mobility separation before ion detection, enabling the visualization of metabolites without the interference of matrix ions. Additional difficulties surrounding low weight metabolite visualization include high resolution imaging, while maintaining sufficient ion numbers for broad and representative analysis of the tissue chemical complement. Here, we use MALDI timsTOF IMS to image low molecular weight metabolites at higher spatial resolution than most metabolite MALDI IMS experiments (20 µm) while maintaining broad coverage within the human kidney. We demonstrate that trapped ion mobility spectrometry (TIMS) can resolve matrix peaks from metabolite signal and separate both isobaric and isomeric metabolites with different distributions within the kidney. The added ion mobility data dimension dramatically increased the peak capacity for spatial metabolomics experiments. Through this improved sensitivity, we have found >40 low molecular weight metabolites in human kidney tissue such as arginic acid, acetylcarnitine, and choline that localize to the cortex, medulla, and renal pelvis, respectively. Future work will involve further exploring metabolomic profiles of human kidneys as a function of age, gender, and ethnicity.more » « less
-
null (Ed.)Ion mobility spectrometry (IMS) mass spectrometry (MS) centers on the ability to separate gaseous structures by size, charge, shape, and followed by mass-to-charge (m/z). For oligomeric structures, improved separation is hypothesized to be related to the ability to extend structures through repulsive forces between cations electrostatically bonded to the oligomers. Here we show the ability to separate differently branched multiply charged ions of star-branched poly(ethylene glycol) oligomers (up to 2000 Da) regardless of whether formed by electrospray ionization (ESI) charged solution droplets or from charged solid particles produced directly from a surface by matrix-assisted ionization. Detailed structural characterization of isomers of the star-branched compositions was first established using a home-built high-resolution ESI IMS-MS instrument. The doubly charged ions have well-resolved drift times, achieving separation of isomers and also allowing differentiation of star-branched versus linear oligomers. An IMS-MS “snapshot” approach allows visualization of architectural dispersity and (im)purity of samples in a straightforward manner. Analyses capabilities are shown for different cations and ionization methods using commercially available traveling wave IMS-MS instruments. Analyses directly from surfaces using the new ionization processes are, because of the multiply charging, not only associated with the benefits of improved gas-phase separations, relative to that of ions produced by matrix-assisted laser desorption/ionization, but also provide the potential for spatially resolved measurements relative to ESI and other ionization methods.more » « less
