CoChat

A Practical Performance Benchmark of Post-Quantum Cryptography Across Heterogeneous Computing Environments

Open access

The DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index; 5 of the 17 sources its claims cite were checked for support.

Abstract

The emergence of large-scale quantum computing presents an imminent threat to contemporary public-key cryptosystems, with quantum algorithms such as Shor’s algorithm capable of efficiently breaking RSA and elliptic curve cryptography (ECC). This vulnerability has catalyzed accelerated standardization efforts for post-quantum cryptography (PQC) by the U.S. National Institute of Standards and Technology (NIST) and global security stakeholders. While theoretical security analysis of these quantum-resistant algorithms has advanced considerably, comprehensive real-world performance benchmarks spanning diverse computing environments—from high-performance cloud infrastructure to severely resource-constrained IoT devices—remain insufficient for informed deployment planning. This paper presents the most extensive cross-platform empirical evaluation to date of NIST-selected PQC algorithms, including CRYSTALS-Kyber and NTRU for key encapsulation mechanisms (KEMs), alongside BIKE as a code-based alternative, and CRYSTALS-Dilithium and Falcon for digital signatures. Our systematic benchmarking framework measures computational latency, memory utilization, key sizes, and protocol overhead across multiple security levels (NIST Levels 1, 3, and 5) in three distinct hardware environments and various network conditions. Results demonstrate that contemporary server architectures can implement these algorithms with negligible performance impact (<5% additional latency), making immediate adoption feasible for cloud services. In contrast, resource-constrained devices experience more significant overhead, with computational demands varying by up to 12× between algorithms at equivalent security levels, highlighting the importance of algorithm selection for edge deployments. Beyond standalone algorithm performance, we analyze integration challenges within existing security protocols, revealing that naive implementation of PQC in TLS 1.3 can increase handshake size by up to 7× compared to classical approaches. To address this, we propose and evaluate three optimization strategies that reduce bandwidth requirements by 40–60% without compromising security guarantees. Our investigation further encompasses memory-constrained implementation techniques, side-channel resistance measures, and hybrid classical-quantum approaches for transitional deployments. Based on these comprehensive findings, we present a risk-based migration framework and algorithm selection guidelines tailored to specific use cases, including financial transactions, secure firmware updates, vehicle-to-infrastructure communications, and IoT fleet management. This practical roadmap enables organizations to strategically prioritize systems for quantum-resistant upgrades based on data sensitivity, resource constraints, and technical feasibility. Our results conclusively demonstrate that PQC is deployment-ready for most applications, provided that implementations are carefully optimized for the specific performance characteristics and security requirements of target environments. We also identify several remaining research challenges for the community, including further optimization for ultra-constrained devices, standardization of hybrid schemes, and hardware acceleration opportunities.

Overview

CoChat Summary

## TL;DR

This paper benchmarks five post-quantum cryptography algorithms—Kyber, NTRU, BIKE, Dilithium, and Falcon—across a server, laptop, and IoT-like device, measuring speed, memory, energy, and TLS overhead. It concludes that PQC is practical across environments, with Kyber generally best for KEMs, Falcon best for verification-heavy signatures, and BIKE the most resource-intensive.

## Key findings

  • On Server E1, all operations complete in ≤ 0.28 ms across algorithms and security levels; the paper says this implies negligible overhead for cloud/data-center workloads.
  • At NIST Security Level 3 on Device E3, Kyber is fastest among KEMs: 1.18 ms key generation, 0.90 ms encryption, and 1.12 ms decryption; BIKE is slowest: 3.05 ms, 1.85 ms, and 2.75 ms respectively.
  • TLS 1.3 handshakes with PQC KEMs stay under 3 ms: Kyber 2.2 ms, NTRU 2.4 ms, BIKE 2.8 ms; classical ECDHE is 1.8 ms.
  • Memory at NIST Level 3 on Device E3 peaks at 850 KB for Kyber, 900 KB for NTRU, 1250 KB for BIKE, 980 KB for Dilithium, and 1250 KB for Falcon.
  • The paper reports NIST Level 3 results are 10–25% better than NIST benchmark values on Server E1, and says PQC key generation is over 1000x faster than RSA for Kyber.

## Methods

  • The study benchmarks five algorithms: CRYSTALS-Kyber, NTRU, BIKE, CRYSTALS-Dilithium, and Falcon, across three environments: a server (E1), laptop (E2), and constrained device (E3).
  • It measures operation latency, key/ciphertext/signature sizes, peak RSS memory, TLS handshake overhead, fragmentation under MTU settings (1500, 512, 256 bytes), concurrency scaling (1 to 16 threads), and energy consumption on Device E3.
  • Each operation was run 1000 times; analysis used arithmetic mean, median, standard deviation, 95th percentile, outlier handling, paired t-tests, bootstrapping, and validation against NIST-published benchmarks.

## Limitations

  • The paper itself notes gaps in prior work rather than its own, but its design still leaves evident limits: it tests only three hardware platforms and only the algorithms listed, so results may not generalize to other devices or PQC schemes (evident from the study scope).
  • It focuses on one implementation stack (liboqs 0.7.2 and OpenSSL-OQS 1.1.1m-OQS), so findings may depend on those software versions and compiler settings (evident from the methodology).
  • The text says the study measured estimated energy consumption on Device E3; it does not state whether energy was measured across all platforms or with hardware power instrumentation beyond that device (stated in the paper).

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/cryptography9020032

Cites

19 of 19 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

5 Valid · 7 Not valid · 6 Not checked · 141 Own results and interpretations

83 more claims

Create a free account to see all 159 claims and to verify the 2 references still open.

See all claims — free account
  • Probably not a substantial claim. 33
  • It may state two separate claims as one. 40
  • The quote only partly states this claim. 3

Create a free account to see these claims.

Citations

Cites

Research works printed but not linked: 18

  1. Chen, M., Yang, F., and Guo, J. (2022, January 2–5). Post-Quantum Cryptography on IoT Devices: Performance and Security Implications. Proceedings of the IEEE International Conference on Computer Communications (INFOCOM), Virtual.
  2. Hamburg, M. (2022, January 13–15). Post-quantum key-exchange for the Internet: Analysis of performance and security. Proceedings of the Real World Crypto Symposium, Amsterdam, The Netherlands.
  3. Abdulrahman, A., Chen, J.P., Chen, Y.J., Hwang, V., Kannwischer, M.J., and Yang, B.Y. (2022, January 13–16). Faster Kyber on Arm Cortex-M4 with NEON. Proceedings of the International Conference on Cryptology and Network Security, Dubai, United Arab Emirates.
  4. Pornin, 2021, Efficient Implementation of Falcon Signatures, IACR Cryptol. ePrint Arch., 2021, 506
  5. Sendrier, 2022, BIKE-2: A New BIKE Design with Enhanced Decoding, IACR Cryptol. ePrint Arch., 2022, 187
  6. Kannwischer, M.J., Rijneveld, J., Schwabe, P., and Stoffelen, K. (2019, January 3–5). PQM4: Post-quantum crypto library for the ARM Cortex-M4. Proceedings of the Constructive Side-Channel Analysis and Secure Design: 10th International Workshop, Darmstadt, Germany.
  7. Mera, 2020, Energy analysis of post-quantum cryptography algorithms on IoT devices, IEEE Internet Things J., 8, 1020
  8. Wang, 2023, Ultra-Low-Memory Post-Quantum Cryptography on Microcontrollers, IEEE Trans. Inf. Forensics Secur., 18, 1631
  9. Kudinov, M., Kuzovkova, A., Kuzovkov, A., and Useinov, R. (2022, January 14–23). Post-Quantum Cryptography Acceleration for IoT: An FPGA-based CRYSTALS-Kyber Accelerator. Proceedings of the Design, Automation & Test in Europe Conference (DATE), Antwerp, Belgium.
  10. Sikeridis, D., Kampanakis, P., and Devetsikiotis, M. (2021, January 14–23). Evaluating the performance of post-quantum TLS. Proceedings of the IEEE International Conference on Communications, Montreal, QC, Canada.
  11. Kwiatkowski, K., and Valenta, L. (2019, January 22–25). Towards Post-Quantum Cryptography in TLS. Proceedings of the The NIST Second PQC Standardization Conference, University of California, Santa Barbara, CA, USA.
  12. Langley, A., and Hamburg, M. (2019, January 24–27). Post-quantum cryptography in TLS 1.3. Proceedings of the Network and Distributed Systems Security (NDSS) Symposium, San Diego, CA, USA.
  13. Crockett, 2023, A Scalability Assessment of Post-Quantum TLS in Cloud Environments, IACR Cryptol. ePrint Arch., 2023, 318
  14. Drucker, N., and Gueron, S. (2022, January 23–25). Performance Analysis of Post-quantum TLS 1.3 on Constrained Devices. Proceedings of the Selected Areas in Cryptography—SAC 2021, Taipei, Taiwan.
  15. Fouque, 2022, Side-Channel Attacks on Post-Quantum Cryptographic Schemes, IACR Trans. Cryptogr. Hardw. Embed. Syst., 2022, 1
  16. Xu, Z., Pemberton, O., and Achutha, R. (2022, January 18–21). A Systematic Evaluation of the Security of Post-Quantum Cryptography Against Side-Channel Attacks. Proceedings of the Cryptographic Hardware and Embedded Systems—CHES 2022, Leuven, Belgium.
  17. Basu, K., Soni, D., Nabeel, M., and Karri, R. (2021, January 1–5). FPGA Acceleration of Post-Quantum Cryptographic Algorithms. Proceedings of the Design, Automation & Test in Europe Conference (DATE), Virtual.
  18. Banerjee, U., Pathak, A., and Chandrakasan, A.P. (June, January 27). Accelerating Post-Quantum Cryptography Through Specialized Hardware for Integer Polynomial Multiplication. Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS), Austin, TX, USA.

Cited by

See all 76

Metadata

Authors
Abbasi, M., Cardoso, F., Váz, P., Silva, J., & Martins, P.
Affiliations
  • Universidade Politécnica de Coimbra (PT)
  • ISLA Santarém - Instituto Politécnico (PT)
  • Universidade Politécnica de Viseu (PT)
Corresponding author
Martins, Pedro
Venue
Cryptography
Publisher
MDPI AG
Published
2025-05-21
Volume
9
Issue
2
Pages
32
ISSN
2410-387X (ISSN-L)
OpenAlex
W4410581134
Cited by
76
References
16
Citations by year
2026: 57 · 2025: 19
FWCI
30.24
Type
journal article
Language
en
Open access
Locations
  • Cryptography (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) (submittedVersion, other-oa)
Topics
  • Physical Sciences › Computer Science › Computer Networks and Communications › Cryptography and Data Security
  • Physical Sciences › Computer Science › Information Systems › Cryptography and Residue Arithmetic
  • Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Quantum Information and Cryptography
Keywords
performance benchmarking, NIST PQC standardization, CRYSTALS-Kyber, NTRU, BIKE, CRYSTALS-Dilithium, falcons, key encapsulation mechanism, digital signatures, resource-constrained devices, Internet of Things security, TLS 1.3, side-channel attack resistance, cloud deployment, edge deployment

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-09-30, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-09-30 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.