The Impossibility of Simultaneous Time and I/O Optimality for The Planar Maxima and Convex Hull Problems
Pith reviewed 2026-05-21 08:49 UTC · model grok-4.3
The pith
No deterministic output-sensitive algorithm can achieve both optimal time and optimal I/O for planar maxima and convex hull.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
We prove that no deterministic output-sensitive algorithm for the planar convex hull and maxima problems can obtain both optimal time and I/O complexity, where the optimality is defined with respect to both the input and output sizes. This explains why the best previous algorithms achieved an optimal I/O bound at the cost of sub-optimal running time. Our results imply that no optimal deterministic output-sensitive cache-oblivious algorithm exists for either problem.
What carries the argument
Lower bound construction in the external-memory model forcing a necessary trade-off between time and I/Os for deterministic algorithms.
If this is right
- This implies no optimal deterministic output-sensitive cache-oblivious algorithm exists for these problems.
- Simple deterministic algorithms exist that match the lower bounds with a trade-off between time and I/Os.
- A randomized version achieves optimal worst-case time and optimal expected I/O bounds.
Where Pith is reading between the lines
- Randomization allows bypassing the impossibility for expected I/O performance.
- The trade-off may be relevant for designing practical algorithms in memory-constrained environments.
- Similar impossibilities could hold for other geometric problems under external memory constraints.
Load-bearing premise
Algorithms are required to be deterministic and output-sensitive under the standard external-memory model with block size B and internal memory M.
What would settle it
Discovery of a deterministic output-sensitive algorithm achieving both the optimal time bound and the optimal I/O bound for these problems would falsify the result.
read the original abstract
We prove that no deterministic output-sensitive algorithm for the planar convex hull and maxima problems can obtain both optimal time and I/O complexity, where the optimality is defined with respect to both the input and output sizes. This explains why the best previous algorithms achieved an optimal I/O bound at the cost of sub-optimal running time (Goodrich et al. [FOCS, 1993]). To the best of our knowledge, the impossibility of simultaneous optimality was only shown previously for the permutation problem by Brodal and Fagerberg [STOC, 2003]. Our results imply that no optimal deterministic output-sensitive cache-oblivious algorithm exists for either problem. In addition, we present simple deterministic algorithms that match our lower bounds and that provide a trade-off between time and I/Os. On the other hand, a simple modification of our deterministic algorithm results in a randomized algorithm that simultaneously achieves optimal (worst-case) time and optimal expected I/O bounds.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proves that no deterministic output-sensitive algorithm for the planar convex hull and maxima problems can simultaneously achieve both optimal time and optimal I/O complexity in the external-memory model (with parameters M and B), where optimality is defined with respect to input size n and output size h. It extends the permutation lower bound of Brodal and Fagerberg, provides deterministic algorithms achieving explicit time-I/O trade-offs that match the lower bound, and gives a simple randomized algorithm achieving optimal worst-case time and optimal expected I/O. The result also implies the non-existence of optimal deterministic cache-oblivious algorithms for these problems.
Significance. If the lower-bound argument holds, this is a significant contribution to external-memory geometric algorithms. It explains the sub-optimal time in prior I/O-optimal algorithms (e.g., Goodrich et al.) and provides the first impossibility result of this form beyond permutations. The matching deterministic trade-off algorithms and the randomized solution that achieves both optima are concrete strengths. The work supplies falsifiable predictions via the explicit trade-off curves and is grounded in the standard external-memory model without hidden assumptions on comparison-based computation or specific h regimes.
major comments (1)
- Lower-bound construction (Section 3): the extension of the Brodal-Fagerberg permutation argument to output-sensitive maxima/convex-hull instances must explicitly verify that the adversary can force Ω((n/B) log_{M/B} (n/h)) I/Os even when the algorithm's running time is allowed to depend on the (unknown) output size h; if the construction permits the algorithm to discover h too early, the I/O lower bound may become trivial for small h.
minor comments (3)
- Abstract: the statement that the deterministic algorithms 'match our lower bounds' would be clearer if it briefly indicated the precise form of the achieved trade-off (e.g., the functional dependence of I/Os on h).
- Introduction: a one-sentence recap of the I/O and time bounds achieved by Goodrich et al. (FOCS 1993) would help readers place the new impossibility result.
- Randomized algorithm section: the analysis of expected I/O optimality should include a short remark on whether the high-probability bound or only expectation is obtained, to avoid ambiguity with the deterministic worst-case time claim.
Simulated Author's Rebuttal
We thank the referee for the careful reading, positive evaluation, and recommendation for minor revision. We address the single major comment below.
read point-by-point responses
-
Referee: Lower-bound construction (Section 3): the extension of the Brodal-Fagerberg permutation argument to output-sensitive maxima/convex-hull instances must explicitly verify that the adversary can force Ω((n/B) log_{M/B} (n/h)) I/Os even when the algorithm's running time is allowed to depend on the (unknown) output size h; if the construction permits the algorithm to discover h too early, the I/O lower bound may become trivial for small h.
Authors: We appreciate this observation. Our adversary argument in Section 3 is constructed so that, for any deterministic algorithm whose running time may depend on the final (unknown) h, the adversary maintains a family of consistent input configurations with different possible output sizes. Resolving which configuration is the true one—and thereby determining the exact h—requires the algorithm to perform Ω((n/B) log_{M/B} (n/h)) I/Os in the worst case before it can safely output a correct maxima set of size h. The construction ensures that any early termination or early inference of h would violate output-sensitivity on at least one of the adversary’s remaining configurations. Consequently the I/O lower bound remains non-trivial even for small h. We will add a short clarifying paragraph after the main adversary lemma to make this verification explicit. revision: yes
Circularity Check
No significant circularity detected in lower-bound proof
full rationale
The manuscript presents an independent lower-bound construction for deterministic output-sensitive algorithms in the standard external-memory model (parameters M and B) for the planar maxima and convex hull problems. It extends the permutation lower bound of Brodal and Fagerberg (STOC 2003) by preserving output sensitivity with respect to both n and h, but the extension itself is developed from first principles within this paper rather than reducing to the prior result by definition or fit. Matching deterministic upper bounds and a randomized variant are supplied separately. No self-definitional equations, fitted inputs renamed as predictions, or load-bearing self-citations that collapse the central impossibility claim were identified. The derivation remains self-contained against the explicitly stated model and determinism assumptions.
Axiom & Free-Parameter Ledger
axioms (1)
- domain assumption Standard external-memory model with parameters M (internal memory size) and B (block size)
Lean theorems connected to this paper
-
IndisputableMonolith/Cost/FunctionalEquation.leanwashburn_uniqueness_aczel unclear?
unclearRelation between the paper passage and the cited Recognition theorem.
We prove that no deterministic output-sensitive algorithm ... can obtain both optimal time ... and optimal I/O complexity ... lower bound ... N·A_s(H) comparisons ... Ω(n(α_1(min{H,√(N/M)})−s)) I/Os
-
IndisputableMonolith/Foundation/ArithmeticFromLogic.leanLogicNat induction / embed_strictMono unclear?
unclearRelation between the paper passage and the cited Recognition theorem.
adversarial strategy ... potential sequence ... Φ(h; (t,κ),S) ... top nodes ... A_s+2ζ−1(1)
What do these tags mean?
- matches
- The paper's claim is directly supported by a theorem in the formal canon.
- supports
- The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
- extends
- The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
- uses
- The paper appears to rely on the theorem as machinery.
- contradicts
- The paper's claim conflicts with a theorem or certificate in the canon.
- unclear
- Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.
Reference graph
Works this paper leans on
-
[1]
1 Wilhelm Ackermann. Zum Hilbertschen Aufbau der reellen Zahlen.Mathematische Annalen, 99:118–133, February 1928.doi:10.1007/BF01459088. 26 The Planar Maxima and Convex Hull Problems 2 Peyman Afshani, Gerth Stølting Brodal, and Nodari Sitchinava. The impossibility of simul- taneous time and I/O optimality for the planar maxima and convex hull problems. In...
-
[2]
3 Peyman Afshani and Pingan Cheng
Schloss Dagstuhl – Leibniz-Zentrum für Informatik. 3 Peyman Afshani and Pingan Cheng. Lower bounds for intersection reporting among flat objects. In Erin W. Chambers and Joachim Gudmundsson, editors,39th International Symposium on Computational Geometry, SoCG 2023, June 12-15, 2023, Dallas, Texas, USA, volume 258 ofLIPIcs, pages 3:1–3:16. Schloss Dagstuhl...
work page 2023
-
[3]
4 Peyman Afshani and Pingan Cheng
doi:10.4230/LIPICS.SOCG.2023.3. 4 Peyman Afshani and Pingan Cheng. On semialgebraic range reporting.Discretete Computa- tional Geometry, 71(1):4–39, 2024.doi:10.1007/S00454-023-00574-1. 5 Peyman Afshani and Arash Farzan. Cache-oblivious output-sensitive two-dimensional convex hull. In Prosenjit Bose, editor,Proceedings of the 19th Annual Canadian Conferen...
-
[4]
URL:http://cccg.ca/ proceedings/2007/07a3.pdf. 6 Peyman Afshani, Chris H. Hamilton, and Norbert Zeh. Cache-oblivious range reporting with optimal queries requires superlinear space. In John Hershberger and Efi Fogel, editors, Proceedings of the 25th ACM Symposium on Computational Geometry, Aarhus, Denmark, June 8-10, 2009, pages 277–286. ACM, 2009.doi:10....
-
[5]
doi:10.1007/S00454-011-9347-7. 8 Alok Aggarwal and S. Vitter, Jeffrey. The input/output complexity of sorting and related problems.Commun. ACM, 31(9):1116–1127, September 1988.doi:10.1145/48529.48535. 9 A.M. Andrew. Another efficient algorithm for convex hulls in two dimensions.Information Processing Letters, 9(5):216–219, 1979.doi:10.1016/0020-0190(79)90...
-
[6]
doi: 10.1016/0020-0190(81)90005-3. P. Afshani, G. S. Brodal, N. Sitchinava 27 16 Gerth Stølting Brodal and Rolf Fagerberg. On the limits of cache-obliviousness. In Lawrence L. Larmore and Michel X. Goemans, editors,Proceedings of the 35th Annual ACM Symposium on Theory of Computing, pages 307–315. ACM, 2003.doi:10.1145/780542.780589. 17 R. C. Buck. Mathem...
-
[7]
Association for Computing Machinery.doi:10.1145/237218.237397. 19 Bernard Chazelle. A minimum spanning tree algorithm with inverse-Ackermann type com- plexity.J. ACM, 47(6):1028–1047, 2000.doi:10.1145/355541.355562. 20 Bernard Chazelle and Burton Rosenberg. The complexity of computing partial sums off- line.International Journal of Computational Geometry ...
-
[8]
doi:10.1142/S0218195991000049. 21 Thomas H. Cormen, Clifford Stein, Ronald L. Rivest, and Charles E. Leiserson.Introduction to Algorithms. The MIT Press, 4nd edition,
-
[9]
22 Rex A. Dwyer. On the convex hull of random points in a polytope.Journal of Applied Probability, 25(4):688–699, 1988.doi:10.2307/3214289. 23 Arash Farzan. Cache-oblivious searching and sorting in multisets. Master’s thesis, University of Waterloo,
-
[10]
Association for Computing Machinery.doi: 10.1145/73007.73040. 25 Matteo Frigo, Charles E. Leiserson, Harald Prokop, and Sridhar Ramachandran. Cache- oblivious algorithms. In40th Annual Symposium on Foundations of Computer Science, FOCS ’99, pages 285–298. IEEE Computer Society, 1999.doi:10.1109/SFFCS.1999.814600. 26 Matteo Frigo, Charles E. Leiserson, Har...
-
[11]
doi:10.1145/2071379. 2071383. 27 Michael T. Goodrich, Jyh-Jong Tsay, Darren Erik Vengroff, and Jeffrey Scott Vitter. External- memory computational geometry. In34th Annual Symposium on Foundations of Computer Science, pages 714–723. IEEE Computer Society, 1993.doi:10.1109/SFCS.1993.366816. 28 R.L. Graham. An efficient algorithm for determining the convex ...
-
[12]
doi:10.1016/ 0020-0190(80)90064-2. 40 Jeffrey Scott Vitter. External memory algorithms and data structures: Dealing with massive data.ACM Computing surveys (CSUR), 33(2):209–271, 2001.doi:10.1145/384192.384193. 41 Jeffrey Scott Vitter. Algorithms and data structures for external memory.Found. Trends Theor. Comput. Sci., 2(4):305–474, January 2008.doi:10.1...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.