pith. machine review for the scientific record. sign in

arxiv: 2605.13766 · v1 · submitted 2026-05-13 · 💻 cs.CE · cs.NA· math.NA· physics.comp-ph

Recognition: unknown

Elastica++: A high-performance, multiphysics framework for large interacting assemblies of Cosserat rods

Tejaswin Parthasarathy , Seung Hyun Kim , Songyuan Cui , Mattia Gazzola
Authors on Pith no claims yet

Pith reviewed 2026-05-14 17:51 UTC · model grok-4.3

classification 💻 cs.CE cs.NAmath.NAphysics.comp-ph
keywords elasticainteractingsoftassembliesdynamicsframeworkhigh-performancelarge
0
0 comments X

The pith

Elastica++ is an open-source high-performance framework implementing the Cosserat rod model for large-scale multiphysics simulations of slender elastic structures.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents Elastica++, a computer program built to simulate how thin, bendy structures such as rods or filaments move when they push, twist, or slide against each other. It relies on the Cosserat rod model, a standard way to track bending, twisting, stretching, and shear in slender bodies. The main technical advance is rewriting the simulation kernels for speed and using shared-memory parallel processing so that thousands of rods can be simulated at once while still handling complex contacts and external forces. The code is also designed to connect easily to other solvers for fluids, magnetism, or active forces. The authors show example runs on nest-like metamaterials, groups of swimming robots, carpets of moving cilia, and magnetic microrobots.

Core claim

Elastica++ combines performance-oriented kernels with shared-memory parallelism to sustain teraflop-scale throughput despite complex discretization domains and physical interactions.

Load-bearing premise

The Cosserat rod model plus the chosen contact and multiphysics coupling schemes remain sufficiently accurate for the demonstrated biophysical regimes without requiring additional validation or model extensions.

Figures

Figures reproduced from arXiv: 2605.13766 by Mattia Gazzola, Seung Hyun Kim, Songyuan Cui, Tejaswin Parthasarathy.

Figure 1
Figure 1. Figure 1: Performance improvements in Elastica++ via various HPC strategies. (a) Ablation study on different processors: relative speed-ups on a single processor obtained cumulatively via (i) SoA2AoS, (ii) explicit (programmer-defined) SIMD, and (iii) data aggregation and blocking, compared against the unoptimized baseline. (b) Roofline analysis of the in-place SO(3) rotation kernel with successive application of op… view at source ↗
Figure 2
Figure 2. Figure 2: Fibrous granular materials. (a) Physical setup of a random packing of 1536 filaments (length L, aspect ratio AR = 31) in a top-open cuboidal enclosure of size 3.34L. (b) A piston, modeled as a moving solid boundary, cyclically compresses the packing. (c) Bulk mechanical responses depicted by quasi-static stress-strain cycles for simulations and control experiments. (d) Execution time distribution for singl… view at source ↗
Figure 3
Figure 3. Figure 3: Active matter. (a) Physical setup illustrating initial random packing of sticks in a top-open cuboidal enclosure. (b) Evolution of scalar nematic order parameter S(t),S¯(t) for the system with 1024, 4096, 16384 filaments. (c) Polar probability distribution averaged over the 125th time period. (d) Simulation snapshots of rod clustering aggregation at different scales: 1024, 4096, and 16384 filaments, at t T… view at source ↗
Figure 4
Figure 4. Figure 4: Magnetized filament assemblies. (a) Experiments6 and one-to-one simulations of an 8×8 carpet of magnetized cilia. (b) Strong scaling of simulation step time with processor count. (c, d) Up-scaled carpet of ∼ 100,000 cilia forming the University of Illinois Wordmark. (e) The metachronal wave pattern at scale. (f, g) Experimental6 and numerical realization of crawling millipede robots, driven magnetically vi… view at source ↗
Figure 5
Figure 5. Figure 5: Schooling of anguilliform swimmers. (a) Simulation setup of 32 anguilliform swimmers arranged in a uniformly-spaced unstructured lattice structure, immersed in a viscous, quiescent fluid (Rea := L 2/(Tνf) = 7143, where νf is the fluid kinematic viscosity.). (b) The motion of a single swimmer at several time instants, colored by instantaneous velocity magnitudes. (c) Snapshot of the schooling swimmers and t… view at source ↗
read the original abstract

Soft, slender structures are ubiquitous in natural and engineered systems, with broad application potential from biomimetic materials to soft robotics. However, there is a notable lack of computational tools that simultaneously preserve high-fidelity continuum rod mechanics, scale to large interacting ensembles, and remain flexible across diverse biophysical settings. Here we introduce Elastica++, an open-source, high-performance implementation of the Cosserat-rod model for large-scale simulations of slender-body dynamics. Elastica++ combines performance-oriented kernels with shared-memory parallelism to sustain teraflop-scale throughput despite complex discretization domains and physical interactions. The framework further interoperates with external numerical solvers, supporting efficient multiphysics workflows. We demonstrate robustness and breadth through case studies spanning passive nest-like metamaterials, collective active-matter dynamics, cilia carpets, soft magnetic microrobots, and schooling swimmers. Elastica++ thus provides a missing foundation for high-throughput studies of emergent behavior in interacting assemblies of elastic slender structures.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript introduces Elastica++, an open-source high-performance implementation of the Cosserat-rod model for large-scale simulations of interacting slender structures. It claims that performance-oriented kernels combined with shared-memory parallelism sustain teraflop-scale throughput even with complex discretization domains and physical interactions, while also supporting multiphysics coupling via external solvers. Robustness is asserted through five case studies covering passive nest-like metamaterials, collective active-matter dynamics, cilia carpets, soft magnetic microrobots, and schooling swimmers.

Significance. If the performance and accuracy claims hold, the work would supply a valuable open-source foundation for high-throughput studies of emergent behavior in elastic slender-body systems, addressing a documented gap in tools that preserve continuum fidelity at scale. The interoperability with external solvers and the breadth of demonstrated biophysical regimes are clear strengths that could accelerate research in soft robotics and biomimetic materials.

major comments (2)
  1. [Abstract] Abstract: the assertion that the framework sustains 'teraflop-scale throughput' despite complex domains and interactions is unsupported by any reported FLOPS counts, roofline analysis, strong-scaling data, or absolute throughput measurements tied to rod count and contact density. This is load-bearing for the central performance claim.
  2. [Case studies] Case studies: robustness is claimed across five passive and active systems, yet no quantitative error metrics, convergence rates under spatial or temporal refinement, or comparisons against known analytical/experimental solutions are supplied. This directly undermines the high-fidelity and accuracy assertions.
minor comments (1)
  1. [Abstract] Abstract: the parallelization strategy (e.g., OpenMP directives, thread affinity, or target hardware) is described only at a high level; a concise statement of the programming model would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive report and for highlighting the need for stronger quantitative support of our performance and accuracy claims. We address each major comment below and will incorporate the requested evidence in a revised manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the assertion that the framework sustains 'teraflop-scale throughput' despite complex domains and interactions is unsupported by any reported FLOPS counts, roofline analysis, strong-scaling data, or absolute throughput measurements tied to rod count and contact density. This is load-bearing for the central performance claim.

    Authors: We agree that the abstract claim requires explicit supporting metrics. In the revision we will add a new performance-analysis subsection (with accompanying figures) that reports measured FLOPS rates, roofline analysis, strong-scaling curves, and absolute throughput numbers explicitly linked to rod count and contact density for the representative cases already presented. These data will be drawn from the existing simulation runs plus targeted additional benchmarks. revision: yes

  2. Referee: [Case studies] Case studies: robustness is claimed across five passive and active systems, yet no quantitative error metrics, convergence rates under spatial or temporal refinement, or comparisons against known analytical/experimental solutions are supplied. This directly undermines the high-fidelity and accuracy assertions.

    Authors: We concur that quantitative validation metrics are essential. The revised manuscript will expand the case-studies section to include (i) L2 or point-wise error norms against analytical solutions or experimental data where available, (ii) spatial and temporal convergence rates under systematic refinement, and (iii) direct comparisons to reference numerical results for the more complex multiphysics examples. Tables summarizing these metrics will be added for each of the five systems. revision: yes

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the established Cosserat rod continuum model and standard numerical methods for rod discretization and contact; no new free parameters, invented physical entities, or ad-hoc axioms are introduced in the abstract.

axioms (1)
  • domain assumption The Cosserat rod model accurately captures the mechanics of slender elastic structures under the interaction regimes examined.
    Invoked as the foundational model for all simulations described in the abstract.

pith-pipeline@v0.9.0 · 5488 in / 1249 out tokens · 52117 ms · 2026-05-14T17:51:19.346003+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

50 extracted references · 50 canonical work pages

  1. [1]

    Aydin, O.et al.Neuromuscular actuation of biohybrid motile bots.Proc. Natl. Acad. Sci.116, 19841–19847 (2019)

    work page 2019

  2. [2]

    K., Parthasarathy, T

    Zhang, X., Chan, F. K., Parthasarathy, T. & Gazzola, M. Modeling and simulation of complex dynamic musculoskeletal architectures.Nat. communications10, 4825 (2019)

    work page 2019

  3. [3]

    Hansell, M.Bird nests and construction behaviour(Cambridge Univer- sity Press, 2000)

    work page 2000

  4. [4]

    bird nest

    Weiner, N., Bhosale, Y ., Gazzola, M. & King, H. Mechanics of randomly packed filaments—the “bird nest” as meta-material.J. Appl. Phys.127 (2020)

    work page 2020

  5. [5]

    advances6, eabc9323 (2020)

    Dong, X.et al.Bioinspired cilia arrays with programmable nonreciprocal motion and metachronal coordination.Sci. advances6, eabc9323 (2020)

    work page 2020

  6. [6]

    communications11, 2637 (2020)

    Gu, H.et al.Magnetic cilia carpets with programmable metachronal waves.Nat. communications11, 2637 (2020)

    work page 2020

  7. [7]

    Tekinalp, A.et al.Topology, dynamics, and control of a muscle- architected soft arm.Proc. Natl. Acad. Sci.121, e2318769121 (2024)

    work page 2024

  8. [8]

    review letters128, 198003 (2022)

    Bhosale, Y .et al.Micromechanical origin of plasticity and hysteresis in nestlike packings.Phys. review letters128, 198003 (2022)

    work page 2022

  9. [9]

    & Tolley, M

    Rus, D. & Tolley, M. Design, fabrication and control of soft robots. Nature521, 467–75, DOI: 10.1038/nature14543 (2015)

    work page doi:10.1038/nature14543 2015

  10. [10]

    Wegst, U. G. K., Bai, H., Saiz, E., Tomsia, A. P. & Ritchie, R. O. Bioinspired structural materials.Nat. Mater.14, 23–36 (2015)

    work page 2015

  11. [11]

    & van Hecke, M

    Bertoldi, K., Vitelli, V ., Christensen, J. & van Hecke, M. Flexible mechanical metamaterials.Nat. Rev. Mater.2, 17066 (2017)

    work page 2017

  12. [12]

    Stains,et al., Anatomy of STEM Teaching in American Universities: A Snapshot from a Large-Scale Observation Study.Science359(6383), 1468–1470 (2018), doi:10.1126/science

    Rogers, J. A., Someya, T. & Huang, Y . Materials and mechanics for stretchable electronics.Science327, 1603–1607, DOI: 10.1126/science. 1182383 (2010)

    work page doi:10.1126/science 2010

  13. [13]

    & Dario, P

    Cianchetti, M., Laschi, C., Menciassi, A. & Dario, P. Biomedical applications of soft robotics.Nat. Rev. Mater.3, 143–153, DOI: 10. 1038/s41578-018-0022-y (2018)

    work page 2018

  14. [14]

    & Cosserat, F

    Cosserat, E. & Cosserat, F. Théorie des corps déformables.Nature81, 67–67

    work page

  15. [15]

    & Truesdell, C

    Ericksen, J. & Truesdell, C. Exact theory of stress and strain in rods and shells.Arch. for Ration. Mech. Analysis1, 295–323 (1957)

    work page 1957

  16. [16]

    Antman, S. S. Kirchhoff’s problem for nonlinearly elastic rods.Q. applied mathematics32, 221–240 (1974)

    work page 1974

  17. [17]

    Simo, J. C. & Vu-Quoc, L. A three-dimensional finite-strain rod model. part ii: Computational aspects.Comput. methods applied mechanics engineering58, 79–116 (1986). 18.Antman, S. S.Nonlinear problems of elasticity(Springer, 2005)

    work page 1986

  18. [18]

    & Eremeyev, V

    Altenbach, H. & Eremeyev, V . A. Cosserat media. InGeneralized Con- tinua from the Theory to Engineering Applications, 65–130 (Springer, 2013)

    work page 2013

  19. [19]

    & Rucker, C

    Till, J., Aloi, V . & Rucker, C. Real-time dynamics of soft and continuum robots based on cosserat rod models.The Int. J. Robotics Res.38, 723–746 (2019). 8/9

    work page 2019

  20. [20]

    & Peruani, F

    Bär, M., Großmann, R., Heidenreich, S. & Peruani, F. Self-propelled rods: Insights and perspectives for active matter.Annu. Rev. Condens. Matter Phys.11, 441–466 (2020)

    work page 2020

  21. [21]

    Doostmohammadi, A., Ignés-Mullol, J., Yeomans, J. M. & Sagués, F. Active nematics.Nat. communications9, 3246 (2018). 23.Lu, H.et al.A bioinspired multilegged soft millirobot that functions in both dry and wet conditions.Nat. communications9, 3944 (2018)

    work page 2018

  22. [22]

    Shih, C.-H.et al.Hierarchical control and learning of a foraging cybe- roctopus.Adv. Intell. Syst.5, 2300088 (2023). 25.Bathe, K.Finite element procedures(Klaus-Jürgen Bathe, 2006)

    work page 2023

  23. [23]

    Bucalem, M. L. & Bathe, K.-J. Locking behavior of isoparametric curved beam finite elements.Appl. Mech. Rev.48, S25–S29, DOI: 10.1115/1.3005077 (1995)

    work page doi:10.1115/1.3005077 1995

  24. [24]

    Felippa, C. A. & Haugen, B. A unified formulation of small-strain corotational finite elements: I. theory.Comput. Methods Appl. Mech. Eng.194, 2285–2335 (2005)

    work page 2005

  25. [25]

    WebsterIII, R. J. & Jones, B. A. Design and kinematic modeling of constant curvature continuum robots: A review.The Int. J. Robotics Res. 29, 1661–1683, DOI: 10.1177/0278364910368147 (2010)

    work page doi:10.1177/0278364910368147 2010

  26. [26]

    & Seneviratne, L

    Renda, F., Boyer, F., Dias, J. & Seneviratne, L. Discrete cosserat ap- proach for multisection soft manipulator dynamics.IEEE Transactions on Robotics34, 1518–1533, DOI: 10.1109/TRO.2018.2868815 (2018)

    work page doi:10.1109/tro.2018.2868815 2018

  27. [27]

    & Mahadevan, L

    Gazzola, M., Dudte, L., McCormick, A. & Mahadevan, L. Forward and inverse problems in the mechanics of soft filaments.Royal Soc. open science5, 171628 (2018)

    work page 2018

  28. [28]

    Pillow 10.3.0

    Tekinalp, A.et al.Gazzolalab/pyelastica: v0.3.1, DOI: 10.5281/zenodo. 7931429 (2023)

    work page doi:10.5281/zenodo 2023

  29. [29]

    Bhosale, Y ., Tekinalp, A., Cui, S., Chan, F. K. & Gazzola, M. Scalable One-stop Platform for Hydroelastic Things (SOPHT), DOI: 10.5281/ zenodo.7658908 (2023)

    work page 2023

  30. [30]

    Intel® threading building blocks.J

    Pheatt, C. Intel® threading building blocks.J. Comput. Sci. Coll.23, 298–298 (2008)

    work page 2008

  31. [31]

    InPractice and Experience in Advanced Research Computing, 1–8, DOI: 10.1145/ 3437359.3465588 (Association for Computing Machinery, New York, NY , USA, 2021)

    Tatineni, M.et al.Expanse: Computing without boundaries. InPractice and Experience in Advanced Research Computing, 1–8, DOI: 10.1145/ 3437359.3465588 (Association for Computing Machinery, New York, NY , USA, 2021)

    work page arXiv 2021

  32. [32]

    & Andrieu, E

    Poquillon, D., Viguier, B. & Andrieu, E. Experimental data about mechanical behaviour during compression tests for various matted fibres. J. materials science40, 5963–5970 (2005)

    work page 2005

  33. [33]

    & Lixu, G

    Eitz, M. & Lixu, G. Hierarchical spatial hashing for real-time collision detection. InIEEE International Conference on Shape Modeling and Applications 2007 (SMI07), 61–70 (IEEE, 2007)

    work page 2007

  34. [34]

    & Pirker, S

    Berger, R., Kloss, C., Kohlmeyer, A. & Pirker, S. Hybrid parallelization of the liggghts open-source dem code.Powder technology278, 234–247 (2015)

    work page 2015

  35. [35]

    & Regueiro, R

    Yan, B. & Regueiro, R. A. Comparison between pure mpi and hybrid mpi-openmp parallelism for discrete element method (dem) of ellipsoidal and poly-ellipsoidal particles.Comput. Part. Mech.6, 271–295 (2019)

    work page 2019

  36. [36]

    & Negrut, D

    Fang, L., Zhang, R., Vanden Heuvel, C., Serban, R. & Negrut, D. Chrono::gpu: An open-source simulation package for granular dynamics using the discrete element method.Processes9, 1813 (2021)

    work page 2021

  37. [37]

    & Libchaber, A

    Wu, X.-L. & Libchaber, A. Particle diffusion in a quasi-two-dimensional bacterial bath.Phys. review letters84, 3017 (2000)

    work page 2000

  38. [38]

    Banerjee, S., Gardel, M. L. & Schwarz, U. S. The actin cytoskeleton as an active adaptive material.Annu. review condensed matter physics11, 421–439 (2020)

    work page 2020

  39. [39]

    Bisoyi, H. K. & Li, Q. Liquid crystals: versatile self-organized smart soft materials.Chem. reviews122, 4887–4926 (2021)

    work page 2021

  40. [40]

    & Michelin, S

    Vinze, P. & Michelin, S. Self-organization of autophoretic suspensions in confined shear flows.Phys. Rev. Fluids9, 014202 (2024)

    work page 2024

  41. [41]

    E., Elgeti, J

    Isele-Holder, R. E., Elgeti, J. & Gompper, G. Self-propelled worm-like filaments: spontaneous spiral formation, structure, and dynamics.Soft matter11, 7181–7190 (2015)

    work page 2015

  42. [42]

    Anand, S. K. & Singh, S. P. Conformation and dynamics of a self- avoiding active flexible polymer.Phys. Rev. E101, 030501 (2020)

    work page 2020

  43. [43]

    & Xu, G.-K

    He, S.-Q., Yin, X., Liang, D., Chang, Z. & Xu, G.-K. Spontaneous oscillation in collective microswimmers: Insights from a chiral self- propelled rod model.Phys. Rev. E111, 014411 (2025). 47.Wensink, H. & Löwen, H. Aggregation of self-propelled colloidal rods near confining walls.Phys. Rev. E-Statistical, Nonlinear , Soft Matter Phys.78, 031409 (2008)

    work page 2025

  44. [44]

    & Chaté, H

    Ginelli, F., Peruani, F., Bär, M. & Chaté, H. Large-scale collective properties of self-propelled rods.Phys. review letters104, 184502 (2010)

    work page 2010

  45. [45]

    & Onck, P

    Zhang, R., Toonder, J. & Onck, P. Metachronal patterns by magnetically- programmable artificial cilia surfaces for low reynolds number fluid transport and mixing.Soft Matter18, DOI: 10.1039/D1SM01680F (2022)

    work page doi:10.1039/d1sm01680f 2022

  46. [46]

    Tekinalp, A., Bhosale, Y ., Cui, S., Chan, F. K. & Gazzola, M. Self- propelling, soft, and slender structures in fluids: Cosserat rods immersed in the velocity–vorticity formulation of the incompressible navier–stokes equations.Comput. Methods Appl. Mech. Eng.440, 117910 (2025)

    work page 2025

  47. [47]

    K., Tan, D

    Venkiteswaran, V . K., Tan, D. K. & Misra, S. Tandem actuation of legged locomotion and grasping manipulation in soft robots using magnetic fields.Extrem. Mech. Lett.41, 101023 (2020)

    work page 2020

  48. [48]

    Carling, J., Williams, T. L. & Bowtell, G. Self-propelled anguilliform swimming: simultaneous solution of the two-dimensional navier–stokes equations and newton’s laws of motion.J. experimental biology201, 3143–3166 (1998)

    work page 1998

  49. [49]

    & Iollo, A

    Bergmann, M. & Iollo, A. Modeling and simulation of fish-like swim- ming.J. Comput. Phys.230, 329–348 (2011)

    work page 2011

  50. [50]

    Code availability The source code and all simulation scripts are publicly available at https://github.com/GazzolaLab/elasticapp

    Weber, P.et al.Optimal flow sensing for schooling swimmers.Biomimet- ics5, 10 (2020). Code availability The source code and all simulation scripts are publicly available at https://github.com/GazzolaLab/elasticapp. Acknowledgments This study was jointly funded by NSF EFRI C3 SoRo #1830881, NSF CAREER #1846752, NSF ELEMENTS #2209322, ONR MURI #N00014–19–1–...

    work page 2020