Pith. sign in

REVIEW 2 major objections 6 minor 60 references

Kinetic Linear Stability Theory for High-Speed Compressible Flows: A High Performance Computing Framework

T0 review · 2 major / 6 minor · reviewed 2026-07-31 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Inside finite-thickness shocks, non-equilibrium velocity distributions systematically push linear stability spectra toward less stable regions than Maxwellian-based analyses predict.

desk verdict First kinetic LST of finite-thickness shocks with a real EQ/NONEQ spectral shift and a working O(10^5) HPC path; the isolation of “VDF non-equilibrium” is a bit messier than the abstract claims, and BGK/Pr=1 is the main physics caveat. read the letter →

arxiv 2607.27440 v1 pith:HZSKS6SH submitted 2026-07-29 physics.flu-dyn

Irmak Taylan Karpuzcu , Deborah Levin , Vassilis Theofilis This is my paper · ORCID
classification physics.flu-dyn
keywords kineticlinearstabilitynormalshocksBoltzmann-BGKtranslationalnon-equilibriumvelocitydistributionfunctionhigh-MachcompressibleflowparalleleigensolversJacobi-Davidson
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper builds the first kinetic linear stability theory for isolated one-dimensional normal shocks. Instead of closing the problem on continuum variables, it linearizes the Boltzmann–BGK equation about kinetic base flows so both the mean shock and the disturbances live at the level of the velocity distribution function. At low Mach number the spectra recover the familiar stable continuous branches. At Mach 3 and 4, keeping the true non-Maxwellian distributions inside the shock shifts those branches toward weaker damping relative to spectra built from the same macroscopic profiles but forced to local Maxwellians. The authors also supply a parallel eigensolver stack so the resulting matrices—hundreds of thousands of unknowns—can actually be solved. A sympathetic reader cares because high-speed vehicle design often trusts continuum or moment-based stability tools; the work says those tools can miss a kinetic imprint even when the mean density and velocity look well resolved.

What carries the argument

Kinetic linear stability theory (kLST): the Boltzmann–BGK equation is linearized about a BE-BGK base flow, with disturbances posed in reduced distribution functions and macroscopic fields recovered only by velocity-space moments, so the stability operator acts on the VDF itself rather than on a closed continuum system.

What would settle it

Repeat the Mach-3 and Mach-4 calculations with a collision operator that recovers the physical Prandtl number (or with a fully kinetic DSMC-linearized operator) and check whether the systematic shift of continuous branches toward weaker damping disappears or reverses.

Watch

Extended reading notes

Core claim

For argon normal shocks at Mach 3 and 4, eigenspectra constructed from non-equilibrium BE-BGK base velocity distribution functions are systematically less stable than spectra that use the same macroscopic base state but enforce local Maxwellian distributions. Continuum-style closures can therefore miss important spectral changes even when macroscopic shock profiles appear adequate.

Load-bearing premise

The BGK collision model with Prandtl number fixed at one is assumed to capture the collision physics that set the linearized spectrum inside the shock.

Editorial extensions

If this is right

  • Stability conclusions drawn from Navier–Stokes or Maxwellian-reconstructed bases for high-Mach shocks can be optimistic relative to a kinetic treatment.
  • Both the base flow and the perturbation problem must retain micro-velocity structure; using only macroscopic kinetic profiles is not enough.
  • Memory-efficient iterative eigensolvers (Jacobi–Davidson with local ILU) become necessary once velocity grids push unknowns past roughly 10^5.
  • The same operator framework can be extended to multidimensional and reacting shock layers once collision models improve.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the less-stable kinetic shift survives better collision models, transition-prediction tools for hypersonic inlets and control surfaces may need a kinetic correction inside strong shocks.
  • The continuous, densely packed shock spectrum is what makes cheap ILU-based Jacobi–Davidson viable; discrete-mode problems such as Couette may still require exact shift-invert.
  • Comparing EQ versus NONEQ spectra is analogous to comparing a single-temperature Boltzmann emission model against a full non-Boltzmann state-to-state model: the macro state alone hides the stability-relevant microphysics.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 6 minor

Summary. The manuscript formulates a kinetic linear stability theory (kLST) by linearizing the Boltzmann–BGK equation in reduced distribution functions about BE-BGK base flows for isolated 1D argon normal shocks, so the stability operator acts on the VDF rather than a closed continuum system. After verifying the discrete operator against compressible Couette eigenvalues of Zou et al., the authors compute spectra at M∞=1.2, 3.0, and 4.0, comparing constructions that enforce local Maxwellian base VDFs (EQ) against those that retain the non-equilibrium BE-BGK VDFs (NONEQ). At higher Mach they report a systematic shift of continuous branches toward less stable regions when non-equilibrium VDFs are retained, and they support the large discrete systems (up to ~2.8×10^5 unknowns) with a parallel SLEPc/PETSc stack (shift-and-invert Arnoldi+MUMPS and Jacobi–Davidson with block-Jacobi ILU), including fill-in and scaling benchmarks on Frontera.

Significance. If the reported EQ/NONEQ spectral shifts hold under the stated operator definitions, the work is a genuine first step beyond NS/moment-based shock LST and beyond prior linearized-BGK studies confined to channel flows: it keeps bi-modal translational non-equilibrium inside both the base state and the perturbation problem for finite-thickness shocks, and it pushes kinetic modal analysis to M∞=4 with a documented HPC path. Strengths that should be credited include external Couette eigenvalue checks (Appendix A, ~1% on least-stable modes), documented base-flow and spectral mesh studies (Appendix B), an explicit shared-grid EQ vs NONEQ protocol, and quantitative solver/fill-in evidence rather than a methods sketch alone. The main scientific payload is the high-Mach spectral comparison; the solver infrastructure is enabling and carefully reported.

major comments (2)
  1. [§II, Eqs. 17–21; §V] §II (Eqs. 17–21, 23) and §V (EQ vs NONEQ): The central claim attributes the high-M spectral shift to translational non-equilibrium / bi-modal base VDFs. In the BGK linearization, setting g_c=g^e_c and h_c=h^e_c simultaneously (i) replaces the base VDF in streaming and moment couplings and (ii) nulls the explicit residual pieces proportional to (g^e_c−g_c) and (h^e_c−h_c) that appear in Eq. 17 and feed the G_ρ, G_u, … / A_g1 blocks. The manuscript never states which of these channels dominates the shift in Figs. 10–13, nor that the EQ macroscopic field with forced Maxwellian VDFs is not itself a steady BE-BGK shock solution. Please add a short operator-level accounting of what EQ changes, and temper language that equates the EQ construction with “continuum” or pure “Maxwellian streaming” unless a controlled split (or an argument why a split is inconsistent) is given.
  2. [Abstract; §V] Abstract and §V: The abstract concludes that “continuum predictions can miss important changes even when the macroscopic profiles appear well resolved.” The high-M evidence is EQ vs NONEQ kinetic spectra (Figs. 10–13), not a side-by-side LNSE or G&P-based cLST spectrum at M∞=3 and 4 on the same Δ-based nondimensionalization. Duck & Balakumar are cited for low-M continuous branches; that does not by itself underwrite the continuum-miss claim at M=3–4. Either add a continuum LST comparison at least at one high-M case, or revise the abstract/conclusion wording to what is actually shown: within BGK-kLST, retaining non-equilibrium base VDFs shifts spectra relative to a Maxwellian-enforced kinetic operator built from the same macros.
minor comments (6)
  1. [Abstract / §I / §VI] §VI already flags BGK/Pr=1; please mirror that limitation once in the abstract or early §I so readers do not over-read the argon spectra as quantitative real-gas predictions.
  2. [Nomenclature; §II] Notation: ϖ vs ω, A_g1 vs A1, and β as both spanwise wavenumber and G&P viscosity constant appear in close proximity; a single consistency pass would help.
  3. [§IV.C, Fig. 5] Fig. 5 compares an incompressible NS extract (n=387) to a kLST extract for qualitative sparsity only; state explicitly in the caption that dimensions and physics differ so the panel is not read as a like-for-like cost comparison.
  4. [§I; §V] Several encoding artifacts remain in the text (e.g., “shockâ=C™s”, “solverâ=C™s”); clean these in production.
  5. [§V, Table 14] Table 14 / length scale: using G&P Δ while base flows are BE-BGK is practical but should be restated when quoting Re_Δ and Kn_Δ next to BGK profiles so nondimensional spectra are unambiguous.
  6. [§V; Appendix C] Appendix C solver tables are valuable; consider pointing to them from the M=4 JD run in §V so the 281,088-unknown result is reproducible from the main text alone.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: EQ/NONEQ spectral shifts are numerical outputs of independently assembled operators, not fits or definitional identities.

full rationale

The load-bearing chain is: (i) BE-BGK (or G&P) base flows from an independent steady solver; (ii) linearization of Boltzmann–BGK about those base states into a generalized eigenproblem on reduced VDFs; (iii) numerical eigenvalues compared between NONEQ (raw base g_c, h_c) and EQ (force g_c=g^e_c, h_c=h^e_c at fixed macros). Nothing in that chain defines the growth-rate shift by construction, fits a target Im(ω), or imports a uniqueness theorem from the authors. Couette verification is against external Zou et al. eigenvalues; low-M continuous branches recover the known Duck–Balakumar structure as a check, not as a renamed empirical law. Self-citations (prior shock/compression-corner work) motivate the problem but do not force the high-M spectral comparison. Methodological worries about whether EQ nulls residual (g^e_c−g_c) collision pieces as well as bi-modal streaming affect isolation/interpretation of “kinetic effects,” not circularity of the derivation. Score 0; steps empty.

Assumptions & free parameters 3 free parameters · 5 assumptions · 1 invented entities

The claim rests on standard kinetic and spectral-stability machinery plus modeling choices (BGK, reduced 1D2V VDFs, temporal normal modes, algebraic shock mapping). No new physical entity is postulated. Numerical resolution and solver tolerances are chosen for convergence rather than fitted to a target growth rate. The main physics liability is BGK with Pr=1, which the authors treat as a domain modeling assumption, not a free fit.

free parameters (3)
  • Spatial/velocity grid pair (N, Q) per Mach = M-dependent; largest n=281088
    Chosen to converge continuous branches (e.g. N=81,Q=20 at M=1.2; N=81,Q=32 at M=3; N=61,Q=48 at M=4). Affects matrix size and resolved spectrum but is not fitted to a measured growth rate.
  • Algebraic mapping length L and far-field half-width s∞
    Hand-chosen to cluster points in the shock and approximate ±∞ Dirichlet decay; changes discrete operator conditioning.
  • Eigensolver shift σ and ILU fill level / JD subspace sizes = σ~0 or 1e-2; ILU(1); ncv~150–200
    Numerical targets and preconditioner richness (Appendix C) that select interior eigenvalues near the origin; not physical constitutive parameters.
assumptions (5)
  • domain assumption BGK collision operator closes the Boltzmann collision integral with a single relaxation time τ=p/μ and implies Pr=1.
    Used for both base flow and linearization throughout §§II–V; limitations stated in §VI.
  • domain assumption Temporal normal-mode ansatz with real spanwise wavenumber β and complex frequency ω on a 1D base shock, with perturbations in reduced distributions g,h.
    Eqs. (18)–(23); standard modal LST structure transferred to kinetic unknowns.
  • domain assumption Inlet/outlet perturbations vanish (Dirichlet) on a mapped domain large enough that asymptotic decay is enforced.
    §II boundary conditions and algebraic mapping §III.A.
  • standard math Gauss–Hermite quadrature and Chebyshev collocation spectrally represent the reduced kinetic operator for the Mach range studied (up to M=4).
    §III.B–C and convergence appendices; standard spectral kinetic discretization.
  • domain assumption Linear stability of the discrete generalized eigenproblem Aq=ωBq is a meaningful proxy for physical modal behavior of the continuous BE-BGK shock.
    Implicit in all of §V; continuous spectra are interpreted via branch shape rather than isolated discrete eigenvalues.
invented entities (1)
  • kLST operator on reduced VDFs about BE-BGK shocks
    purpose: Name and assemble the stability problem so the unknown is the perturbed distribution rather than a continuum field vector.
    Methodological construct, not a new particle or force; built from standard BGK linearization plus shock base flows.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Kinetic Linear Stability Theory for High-Speed Compressible Flows: A High Performance Computing Framework." pith.science (2026). https://pith.science/paper/HZSKS6SH

@misc{pith2026260727440,
  author       = {Pith},
  title        = {Pith review of: Kinetic Linear Stability Theory for High-Speed Compressible Flows: A High Performance Computing Framework},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HZSKS6SH}},
  note         = {Machine review of arXiv:2607.27440}
}
abstract

Shock waves in high-speed compressible flows contain finite-thickness, high-gradient regions where the continuum assumption becomes questionable and translational non-equilibrium arises, including non-Maxwellian micro-velocity distributions. Classical shock stability analyses rely on Navier-Stokes or moment closures and cannot retain bi-modal velocity distributions inside the shock. We develop and apply, for the first time, a kinetic linear stability theory (kLST) for one-dimensional normal shocks by linearizing the Boltzmann-BGK equation about kinetic BE-BGK base flows. Perturbations are posed in reduced distribution functions, with macroscopic fields recovered by velocity-space moments, so the stability operator acts on the VDF rather than a closed continuum system. Verified against compressible Couette eigenvalue benchmarks near continuum, the framework is applied to argon shocks at $M_\infty=1.2$, $3.0$, and $4.0$. At low Mach number, where BE-BGK and Gilbarg-Paolucci profiles nearly coincide, the spectra recover stable continuous branches. At higher Mach number, comparing Maxwellian and non-equilibrium VDF-based eigenspectra shows that kinetic effects shift the spectrum toward less stable regions, so continuum predictions can miss important changes even when macroscopic profiles appear well resolved. For large high-Mach matrices--$O(10^5)$ unknowns and up to billions of nonzeros--we develop a parallel SLEPc/PETSc infrastructure using shift-and-invert Arnoldi with MUMPS LU for moderate sizes and Jacobi-Davidson (JD) with block-Jacobi ILU for the largest systems. Coupled spatial/micro-velocity sparsity causes severe LU fill-in, making direct solvers memory-limited and motivating JD. We compute kLST spectra for an $M_\infty=4.0$ shock with 281088 unknowns, to our knowledge the highest-Mach kinetic linear stability calculation reported for isolated finite-thickness shock layers.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 28 canonical work pages

  1. [1]

    Anomalous heat transfer and pressure distributions on blunt bodies at hypersonic speeds in the presence of an impinging shock

    Edney, B., “Anomalous heat transfer and pressure distributions on blunt bodies at hypersonic speeds in the presence of an impinging shock.” Tech. Rep. 115, The Aeronautical Research Institute of Sweden, 1968. https://doi.org/10.2172/4480948, URL https://www.osti.gov/biblio/4480948

    arXiv 1968

  2. [2]

    On the synchronisation of three-dimensional shock layer and laminar separation bubble instabilities in hypersonic flow over a double wedge,

    Sawant, S. S., Theofilis, V., and Levin, D., “On the synchronisation of three-dimensional shock layer and laminar separation bubble instabilities in hypersonic flow over a double wedge,”Journal of Fluid Mechanics, Vol. 941, 2022, p. A7. https://doi.org/10.1017/jfm.2022.276

    work page doi:10.1017/jfm.2022.276 2022

  3. [3]

    On linear stability of supersonic flow over a short compression corner at large ramp angles

    Karpuzcu, I. T., Theofilis, V., and Levin, D. A., “On linear stability of supersonic flow over a short compression corner at large ramp angles,”arXiv preprint, 2024. https://doi.org/10.48550/arXiv.2405.06775

    work page Pith review arXiv doi:10.48550/arxiv.2405.06775 2024

  4. [4]

    Electron beam density measurements in shock waves in argon,

    Schmidt, B., “Electron beam density measurements in shock waves in argon,”Journal of Fluid Mechanics, Vol. 39, No. 2, 1969, pp. 361–373. https://doi.org/10.1017/S0022112069002229

    work page doi:10.1017/s0022112069002229 1969

  5. [5]

    Density profiles in argon and nitrogen shock waves measured by the absorption of an electron beam,

    Alsmeyer, H., “Density profiles in argon and nitrogen shock waves measured by the absorption of an electron beam,”Journal of Fluid Mechanics, Vol. 74, No. 3, 1976, pp. 497–513. https://doi.org/10.1017/S0022112076001912

    work page doi:10.1017/s0022112076001912 1976

  6. [6]

    Evaluation of Two Statistical Models Using the Shock Structure Problem,

    Giddens, D. P., Huang, A. B., and Young, V. Y. C., “Evaluation of Two Statistical Models Using the Shock Structure Problem,”The Physics of Fluids, Vol. 14, No. 12, 1971-12, pp. 2645–2651. https://doi.org/10.1063/1.1693387, URL https://doi.org/10.1063/1.1693387

    work page doi:10.1063/1.1693387 1971

  7. [7]

    AModelforCollisionProcessesinGases.I.SmallAmplitudeProcessesinCharged and Neutral One-Component Systems,

    Bhatnagar,P.L.,Gross,E.P.,andKrook,M.,“AModelforCollisionProcessesinGases.I.SmallAmplitudeProcessesinCharged and Neutral One-Component Systems,”Phys. Rev., Vol. 94, 1954-05, pp. 511–525. https://doi.org/10.1103/PhysRev.94.511, URL https://link.aps.org/doi/10.1103/PhysRev.94.511

    doi:10.1103/physrev.94.511 1954

  8. [8]

    Linear stability of the cylindrical Couette flow of a rarefied gas,

    Yoshida, H., and Aoki, K., “Linear stability of the cylindrical Couette flow of a rarefied gas,”Phys. Rev. E, Vol. 73, 2006-02, p. 021201. https://doi.org/10.1103/PhysRevE.73.021201, URL https://link.aps.org/doi/10.1103/PhysRevE.73.021201

    work page doi:10.1103/physreve.73.021201 2006

Show all 60 references
  1. [9]

    Modeling of viscous shock tube using ES-BGK model kinetic equations,

    Chigullapalli, S., Venkattraman, A., and Alexeenko, A., “Modeling of viscous shock tube using ES-BGK model kinetic equations,”47th AIAA Aerospace Sciences Meeting including The New Horizons Forum and Aerospace Exposition, 2009, p. 1317

    2009

  2. [10]

    A.,Molecular Gas Dynamics and the Direct Simulation of Gas Flows, Clarendon, Oxford, England, U.K., 1994

    Bird, G. A.,Molecular Gas Dynamics and the Direct Simulation of Gas Flows, Clarendon, Oxford, England, U.K., 1994

    1994

  3. [11]

    Onset of three-dimensionality in supersonic flow over a slender double wedge,

    Sidharth, G., Dwivedi, A., Candler, G. V., and Nichols, J. W., “Onset of three-dimensionality in supersonic flow over a slender double wedge,”Physical Review Fluids, Vol. 3, No. 9, 2018, p. 093901

    2018

  4. [12]

    Occurrence of global instability in hypersonic compression corner flow,

    Hao, J., Cao, S., Wen, C.-Y., and Olivier, H., “Occurrence of global instability in hypersonic compression corner flow,”Journal of Fluid Mechanics, Vol. 919, 2021, p. A4. https://doi.org/10.1017/jfm.2021.372

    2021 doi

  5. [13]

    On the Stability of Normal Shock Waves,

    Duck, P. W., and Balakumar, P., “On the Stability of Normal Shock Waves,”Instability, Transition, and Turbulence, edited by M. Y. Hussaini, A. Kumar, and C. L. Streett, Springer New York, New York, NY, 1992, pp. 253–265. 61

    1992

  6. [14]

    The Structure of Shock Waves in the Continuum Theory of Fluids,

    Gilbarg, D., and Paolucci, D., “The Structure of Shock Waves in the Continuum Theory of Fluids,”Journal of Rational Mechanics and Analysis, Vol. 2, 1953, pp. 617–642. URL http://www.jstor.org/stable/24900350

    1953

  7. [15]

    The development of kinetic models and simulation methods to study molecular fluctuations, modal response, shock-laminar separation bubble instabilities,

    Sawant, S. S., “The development of kinetic models and simulation methods to study molecular fluctuations, modal response, shock-laminar separation bubble instabilities,” Ph.D. thesis, University of Illinois, Urbana Champaign, 2022

    2022

  8. [16]

    A kinetic approach to studying low-frequency molecular fluctuations in a one-dimensional shock,

    Sawant, S. S., Levin, D. A., and Theofilis, V., “A kinetic approach to studying low-frequency molecular fluctuations in a one-dimensional shock,”Physics of Fluids, Vol. 33, No. 10, 2021-10, p. 104106. https://doi.org/10.1063/5.0065971, URL https://doi.org/10.1063/5.0065971

    2021 doi

  9. [17]

    Stability of shock waves,

    Kuznetsov, N. M., “Stability of shock waves,”Soviet Physics Uspekhi, Vol. 32, No. 11, 1989-11, p. 993. https://doi.org/10. 1070/PU1989v032n11ABEH002777, URL https://dx.doi.org/10.1070/PU1989v032n11ABEH002777

    1989 doi

  10. [18]

    Shock wave stability,

    Swan, G. W., and Fowles, G. R., “Shock wave stability,”The Physics of Fluids, Vol. 18, No. 1, 1975-01, pp. 28–35. https://doi.org/10.1063/1.860989, URL https://doi.org/10.1063/1.860989

    1975 doi

  11. [19]

    Stability of Plane Shock Waves,

    Fowles, G. R., and Swan, G. W., “Stability of Plane Shock Waves,”Phys. Rev. Lett., Vol. 30, 1973-05, pp. 1023–1025. https://doi.org/10.1103/PhysRevLett.30.1023, URL https://link.aps.org/doi/10.1103/PhysRevLett.30.1023

    1973 doi

  12. [20]

    Stability of Step Shocks,

    Erpenbeck, J. J., “Stability of Step Shocks,”The Physics of Fluids, Vol. 5, No. 10, 1962-10, pp. 1181–1187. https: //doi.org/10.1063/1.1706503, URL https://doi.org/10.1063/1.1706503

    1962 doi

  13. [21]

    Reply to Comments by Gardner,

    Erpenbeck, J. J., “Reply to Comments by Gardner,”The Physics of Fluids, Vol. 6, No. 9, 1963-09, pp. 1368–1368. https://doi.org/10.1063/1.1706918, URL https://doi.org/10.1063/1.1706918

    1963 doi

  14. [22]

    Comment on “Stability of Step Shocks

    Gardner, C. S., “Comment on “Stability of Step Shocks”,”The Physics of Fluids, Vol. 6, No. 9, 1963-09, pp. 1366–1367. https://doi.org/10.1063/1.1706917, URL https://doi.org/10.1063/1.1706917

    1963 doi

  15. [23]

    On the Cauchy Problem for Boltzmann Equations: Global Existence and Weak Stability,

    DiPerna, R. J., and Lions, P. L., “On the Cauchy Problem for Boltzmann Equations: Global Existence and Weak Stability,” Annals of Mathematics, Vol. 130, No. 2, 1989, pp. 321–366. https://doi.org/10.2307/1971423

    1989 doi

  16. [24]

    On the Existence of Global Solutions of Mixed Problem for Non-linear Boltzmann Equation,

    Ukai, S., “On the Existence of Global Solutions of Mixed Problem for Non-linear Boltzmann Equation,”Proceedings of the Japan Academy, Vol. 50, No. 3, 1974, pp. 179–184. https://doi.org/10.3792/pja/1195519027

    1974

  17. [25]

    The Vlasov–Maxwell–Boltzmann System Near Maxwellians,

    Guo, Y., “The Vlasov–Maxwell–Boltzmann System Near Maxwellians,”Inventiones Mathematicae, Vol. 153, No. 3, 2003, pp. 593–630. https://doi.org/10.1007/s00222-003-0301-z

    2003 doi

  18. [26]

    A new linear stability analysis approach for microchannel flow based on the Boltzmann Bhatnagar–Gross–Krook equation,

    Zou, S., Zhong, C., Bi, L., Yuan, X., and Tang, Z., “A new linear stability analysis approach for microchannel flow based on the Boltzmann Bhatnagar–Gross–Krook equation,”Physics of Fluids, Vol. 34, No. 12, 2022-12, p. 124114. https://doi.org/10.1063/5.0131135, URL https://doi...

    2022 doi

  19. [27]

    A novel linear stability analysis method for plane Couette flow considering rarefaction effects,

    Zou, S., Bi, L., Zhong, C., Yuan, X., and Tang, Z., “A novel linear stability analysis method for plane Couette flow considering rarefaction effects,”Journal of Fluid Mechanics, Vol. 963, 2023, p. A33. https://doi.org/10.1017/jfm.2023.230. 62

    2023 doi

  20. [28]

    The equivalence between BE-LSE and NS-LSEs under continuum assumption,

    Ma, Q., Lv, J., and Bi, L., “The equivalence between BE-LSE and NS-LSEs under continuum assumption,”Communications in Nonlinear Science and Numerical Simulation, Vol. 128, 2024, p. 107637. https://doi.org/https://doi.org/10.1016/j.cnsns.2023. 107637, URL https://www.sciencedir...

    2024 doi

  21. [29]

    Anormal-modeapproachforhigh-speedrarefiedplaneCouetteflow,

    Zou,S.,Bi,L.,Zhong,C.,Yuan,X.,andTang,Z.,“Anormal-modeapproachforhigh-speedrarefiedplaneCouetteflow,”Physics of Fluids, Vol. 36, No. 1, 2024-01, p. 014121. https://doi.org/10.1063/5.0187318, URL https://doi.org/10.1063/5.0187318

    2024 doi

  22. [30]

    SLEPc: A Scalable and Flexible Toolkit for the Solution of Eigenvalue Problems,

    Hernandez, V., Roman, J. E., and Vidal, V., “SLEPc: A Scalable and Flexible Toolkit for the Solution of Eigenvalue Problems,” ACM Transactions on Mathematical Software, Vol. 31, No. 3, 2005, pp. 351–362

    2005

  23. [31]

    PETSc/TAO Users Manual,

    Balay, S., Abhyankar, S., Adams, M. F., Benson, S., Brown, J., Brune, P., Buschelman, K., Constantinescu, E. M., Dalcin, L., Dener, A., Eijkhout, V., Faibussowitsch, J., Gropp, W. D., Hapla, V., Isaac, T., Jolivet, P., Karpeev, D., Kaushik, D., Knepley, M. G., Kong, F., Kruger...

    2023

  24. [32]

    A Survey of Software for Sparse Eigenvalue Problems,

    Hernández, V., Román, J. E., Tomás, A., and Vidal, V., “A Survey of Software for Sparse Eigenvalue Problems,” SLEPc Technical Report STR-6, Universidad Politécnica de Valencia, 2009. URL http://www.grycap.upv.es/slepc

    2009

  25. [33]

    ImprovingHigh-PerformanceSparseLibrariesUsingCompiler-AssistedSpecialization: A PETSc Case Study,

    Ramalingam,S.,Hall,M.,andChen,C.,“ImprovingHigh-PerformanceSparseLibrariesUsingCompiler-AssistedSpecialization: A PETSc Case Study,”2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum, 2012, pp. 487–496. https://doi.org/10.1109/IPDP...

    2012 doi

  26. [34]

    Solving large sparse linear systems in a grid environment: the GREMLINS code versusthePETSclibrary,

    Jezequel, F., Couturier, R., and Denis, C., “Solving large sparse linear systems in a grid environment: the GREMLINS code versusthePETSclibrary,”TheJournalofSupercomputing,Vol.59,No.3,2012,pp.1517–1532. https://doi.org/10.1007/s11227- 011-0563-y

    2012 doi

  27. [35]

    AParallelImplementationofDavidsonMethodsforLarge-ScaleEigenvalueProblemsinSLEPC,

    Romero,E.,andRoman,J.E.,“AParallelImplementationofDavidsonMethodsforLarge-ScaleEigenvalueProblemsinSLEPC,” ACM Transactions on Mathematical Software (TOMS), Vol. 40, No. 2, 2014, pp. 1–29. https://doi.org/10.1145/2543696

    2014 doi

  28. [36]

    Parallel Krylov Solvers for the Polynomial Eigenvalue Problem in SLEPC,

    Campos, C., and Roman, J. E., “Parallel Krylov Solvers for the Polynomial Eigenvalue Problem in SLEPC,”SIAM Journal on Scientific Computing, Vol. 38, No. 5, 2016, pp. S385–S411. https://doi.org/10.1137/15M1022458

    2016 doi

  29. [37]

    NEP: A Module for the Parallel Solution of Nonlinear Eigenvalue Problems in SLEPC,

    Campos, C., and Roman, J. E., “NEP: A Module for the Parallel Solution of Nonlinear Eigenvalue Problems in SLEPC,”ACM Transactions on Mathematical Software (TOMS), Vol. 47, No. 3, 2021, pp. 1–29. https://doi.org/10.1145/3447544

    2021 doi

  30. [38]

    ComparisonofNumericalMethodsandOpen-SourceLibrariesforEigenvalue AnalysisofLarge-ScalePowerSystems,

    Tzounas,G.,Dassios,I.,Liu,M.,andMilano,F.,“ComparisonofNumericalMethodsandOpen-SourceLibrariesforEigenvalue AnalysisofLarge-ScalePowerSystems,”AppliedSciences,Vol.10,No.21,2020,p.7592. https://doi.org/10.3390/app10217592

    2020 doi

  31. [39]

    Stability of Natural Convection in a Vertical Porous Layer of Viscoelastic Navier-Stokes-Voigt Fluid,

    Shivaraj, D. L., “Stability of Natural Convection in a Vertical Porous Layer of Viscoelastic Navier-Stokes-Voigt Fluid,”ASME Journal of Heat and Mass Transfer, Vol. 147, No. 10, 2025, p. 102601. https://doi.org/10.1115/1.4069013. 63

    2025 doi

  32. [40]

    Resolvent4py: A parallel Python package for analysis, model reduction and control of large-scale linear systems,

    Padovan, A., Anantharaman, V., Rowley, C. W., Vollmer, B., Colonius, T., and Bodony, D. J., “Resolvent4py: A parallel Python package for analysis, model reduction and control of large-scale linear systems,”SoftwareX, Vol. 31, 2025, p. 102286. https://doi.org/10.1016/j.softx.20...

    2025

  33. [41]

    A.,Numerical analysis of spectral methods: theory and applications, SIAM, 1977

    Gottlieb, D., and Orszag, S. A.,Numerical analysis of spectral methods: theory and applications, SIAM, 1977

    1977

  34. [42]

    Y., Quarteroni, A., Zang, T

    Canuto, C., Hussaini, M. Y., Quarteroni, A., Zang, T. A., et al.,Spectral Methods in Fluid Dynamics [electronic resource], Berlin, Heidelberg: Springer Berlin Heidelberg„ 1988

    1988

  35. [43]

    Boyd, J.,Chebyshev and Fourier Spectral Methods: Second Revised Edition, Dover Books on Mathematics, Dover Publications,

  36. [44]

    N.,Spectral Methods in MATLAB, Society for Industrial and Applied Mathematics, 2000

    Trefethen, L. N.,Spectral Methods in MATLAB, Society for Industrial and Applied Mathematics, 2000. https://doi.org/10.1137/ 1.9780898719598, URL https://epubs.siam.org/doi/abs/10.1137/1.9780898719598

    2000 doi

  37. [45]

    c., Imprimerie de l’Académie impériale des sciences, 1853

    Chebyshev, P.,Théorie des mécanismes connus sous le nom de parallélogrammes, 1. c., Imprimerie de l’Académie impériale des sciences, 1853. URL https://books.google.com.tr/books?id=o7AwkgAACAAJ

  38. [46]

    Linear Stability Analysis of Canonical Flows Using Boltzmann BGK Equation,

    Karpuzcu, I. T., and Levin, D., “Linear Stability Analysis of Canonical Flows Using Boltzmann BGK Equation,”AIAA SCITECH 2025 Forum, American Institute of Aeronautics and Astronautics, 2025, pp. AIAA Paper 2025–1313. https: //doi.org/10.2514/6.2025-1313

    2025 doi

  39. [47]

    Towards the Ultimate Conservative Difference Scheme. V. A Second-Order Sequel to Godunov’s Method,

    van Leer, B., “Towards the Ultimate Conservative Difference Scheme. V. A Second-Order Sequel to Godunov’s Method,” Journal of Computational Physics, Vol. 32, No. 1, 1979, pp. 101–136. https://doi.org/10.1016/0021-9991(79)90145-1

    1979 doi

  40. [48]

    An Algorithm for Generalized Matrix Eigenvalue Problems,

    Moler, C. B., and Stewart, G. W., “An Algorithm for Generalized Matrix Eigenvalue Problems,”SIAM Journal on Numerical Analysis, Vol. 10, No. 2, 1973, pp. 241–256. URL http://www.jstor.org/stable/2156353

    1973

  41. [49]

    The principle of minimized iterations in the solution of the matrix eigenvalue problem,

    Arnoldi, W. E., “The principle of minimized iterations in the solution of the matrix eigenvalue problem,”Quarterly of Applied Mathematics, Vol. 9, No. 1, 1951, pp. 17–29. URL http://www.jstor.org/stable/43633863

    1951

  42. [50]

    Solution of Systems of Linear Equations by Minimized Iterations1,

    Lanczos, C., “Solution of Systems of Linear Equations by Minimized Iterations1,”Journal of research of the National Bureau of Standards, Vol. 49, 1952, p. 33. URL https://api.semanticscholar.org/CorpusID:7484650

    1952

  43. [51]

    A Matlab implementation of the Implicitly Restarted Arnoldi Method for solving large-scale eigenvalue problems,

    Radke, R., “A Matlab implementation of the Implicitly Restarted Arnoldi Method for solving large-scale eigenvalue problems,” Ph.D. thesis, Rice University, 1996

    1996

  44. [52]

    Implicitly Restarted Arnoldi/Lanczos Methods for Large Scale Eigenvalue Calculations,

    Sorensen, D. C., “Implicitly Restarted Arnoldi/Lanczos Methods for Large Scale Eigenvalue Calculations,”Parallel Numerical Algorithms, edited by D. E. Keyes, A. Sameh, and V. Venkatakrishnan, Springer Netherlands, Dordrecht, 1997, pp. 119–165. https://doi.org/10.1007/978-94-01...

    1997 doi

  45. [53]

    A Fully Asynchronous Multifrontal Solver Using Distributed Dynamic Scheduling,

    Amestoy, P. R., Duff, I. S., Koster, J., and L’Excellent, J.-Y., “A Fully Asynchronous Multifrontal Solver Using Distributed Dynamic Scheduling,”SIAM Journal on Matrix Analysis and Applications, Vol. 23, No. 1, 2001, pp. 15–41. 64

    2001

  46. [54]

    A Jacobi–Davidson Iteration Method for Linear Eigenvalue Problems,

    Sleijpen, G. L. G., and der Vorst, H. A. V., “A Jacobi–Davidson Iteration Method for Linear Eigenvalue Problems,”SIAM Journal on Matrix Analysis and Applications, Vol. 17, No. 2, 1996, pp. 401–425. https://doi.org/10.1137/S0895479894270427

    1996 doi

  47. [55]

    GMRES:AGeneralizedMinimalResidualAlgorithmforSolvingNonsymmetricLinearSystems,

    Saad,Y.,andSchultz,M.H.,“GMRES:AGeneralizedMinimalResidualAlgorithmforSolvingNonsymmetricLinearSystems,” SIAM Journal on Scientific and Statistical Computing, Vol. 7, No. 3, 1986, pp. 856–869. https://doi.org/10.1137/0907058

    1986 doi

  48. [56]

    A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs,

    Karypis, G., and Kumar, V., “A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs,”SIAM Journal on Scientific Computing, Vol. 20, No. 1, 1998, pp. 359–392

    1998

  49. [57]

    Frontera Supercomputer,

    “Frontera Supercomputer,” https://www.tacc.utexas.edu/systems/frontera, 2022. Accessed: 2022-08-26

    2022

  50. [58]

    Modal Stability Theory: Lecture notes from the FLOW-NORDITA Summer School on Advanced Instability Methods for Complex Flows, Stockholm, Sweden, 2013,

    Juniper, M. P., Hanifi, A., and Theofilis, V., “Modal Stability Theory: Lecture notes from the FLOW-NORDITA Summer School on Advanced Instability Methods for Complex Flows, Stockholm, Sweden, 2013,”Applied Mechanics Reviews, Vol. 66, No. 2, 2014-03, p. 024804. https://doi.org/...

    2013 doi

  51. [59]

    Modeling of Nitric Oxide Vibrational Level Populations for High Mach Number Flows,

    Thirani, S., Karpuzcu, I. T., and Levin, D. A., “Modeling of Nitric Oxide Vibrational Level Populations for High Mach Number Flows,”Journal of Thermophysics and Heat Transfer, Vol. 39, No. 4, 2025, pp. 835–849. https://doi.org/10.2514/1.T7111. 65

    2025 doi

  52. [2013]

    URL https://books.google.com/books?id=b4TCAgAAQBAJ

Pith tools

Reviewed July 31, 2026 · model on record in the stance chip above.