Skip to content

Journal · High-performance computing · Updated 22 November 2026

The Journal of Supercomputing

Read The Journal of Supercomputing articles on DeepDyve. No university library needed.

The Journal of Supercomputing has published since 1987 and covers high-performance computing and the algorithms that run on it, with recent papers on mixed-precision numerics in science, image encryption, machine learning for medical prediction and lightweight object detection. Springer Nature publishes it. A DeepDyve subscription lets you read it in full, alongside millions of articles from other journals licensed from publishers including Springer Nature.

Then $59 a month or $499 a year. A credit card is required; cancel during the trial and you won’t be charged. You read articles online; PDFs are sold separately.

Personal plans aren’t for business use. Reading for work? See Enterprise.

Cover of The Journal of Supercomputing
Journal cover

The Journal of Supercomputing at a glance

Full title
The Journal of Supercomputing
Abbreviation (NLM)
J Supercomput
Publisher
Springer Nature
Print ISSN
0920-8542
Online ISSN (eISSN)
1573-0484
First published
1987
Access model
Hybrid (10% free to read in 2026)

Ways to access The Journal of Supercomputing

10% of the articles The Journal of Supercomputing published from January 1 to November 20, 2026 are open access and free to read. For the rest, look for the authors’ manuscript, check a library, or read them on DeepDyve.

  1. Open access articles. 10% of The Journal of Supercomputing articles published from January 1 to November 20, 2026 carry a Creative Commons license, per Crossref, and are free to read. They’re marked “Open access” below.
  2. The authors’ accepted manuscript. For subscription articles, Springer Nature lets authors post their accepted manuscript on their own website as soon as it’s accepted, and in a repository such as PubMed Central 12 months after publication.
  3. Ask the authors. Authors can also share a free, read-only link to their published article through Springer Nature’s SharedIt, so it’s worth asking the authors directly.
  4. NIH-funded research. For papers accepted on or after July 1, 2025, the US National Institutes of Health requires the accepted manuscript in PubMed Central, free with no embargo from the official publication date.
  5. A library. Your university, hospital or public library may subscribe to The Journal of Supercomputing. Check its journal list and whether it offers remote access.
  6. DeepDyve. Read The Journal of Supercomputing articles online for $59 a month or $499 a year after a free first month. Articles marked LitStream below are in DeepDyve’s LitStream collection, which a subscription lets you read online. PDFs are sold separately, and availability varies by article.

Sources: Springer Nature self-archiving policy; NIH Public Access Policy; open access share from Crossref license data.

Impact factor and metrics

The Journal of Supercomputing’s 2025 Journal Impact Factor is 3.5, and its 5-year Impact Factor is 3.1.

3.5

Journal Impact Factor (2025)

3.1

5-year Impact Factor (2025)

38 days

Median time to first decision

1.3M

Downloads (2025)

Source: journal metrics on the The Journal of Supercomputing homepage (Springer Nature), checked November 22, 2026. Impact Factor from Clarivate Journal Citation Reports.

Recent Journal of Supercomputing research worth reading

Published January to May 2026. Summaries describe what the authors report.

  1. LitStreamImage encryption · Chaotic maps

    Design of a three-dimensional logistic map and its application to seafood image encryption

    The authors propose a reversible encryption method for seafood images built on a new three-dimensional logistic map with a larger key space than the one-dimensional version, and report high security at low computational cost, allowing real-time, high-throughput encryption.

    Ding et al. · The Journal of Supercomputing, published online March 6, 2026 · DOI 10.1007/s11227-025-08196-5

  2. Open accessMixed precision · Scientific computing

    Mixed-precision numerics in scientific applications: survey and perspectives

    The authors survey how fluid dynamics, weather and climate, quantum chemistry and computational genomics have begun using mixed-precision methods such as iterative refinement and adaptive precision solvers, and argue these can speed up science on new hardware without sacrificing accuracy.

    Kashi et al. · The Journal of Supercomputing, published online March 24, 2026 · DOI 10.1007/s11227-026-08264-4

  3. LitStreamMachine learning · Uterine adhesions

    Early predictive modeling of uterine adhesions severity: a comparative study of ANN and ANFIS techniques

    Using 300 clinical cases, the authors build neural network and neuro-fuzzy models to predict how severe uterine adhesions will be after repeated cesarean sections, reporting a test R-value of 0.84802 for the neural network and greater interpretability for the neuro-fuzzy model.

    Elgohr et al. · The Journal of Supercomputing, published online January 4, 2026 · DOI 10.1007/s11227-025-08145-2

  4. LitStreamObject detection · Aerial imagery

    IRT-DETR: a lightweight transformer with ConvNextV2 and Cascaded Group Attention for real-time military aircraft detection in aerial imagery

    The authors build a lightweight detection transformer for spotting small military aircraft in drone images and report that, on a dataset of 3842 images, it cuts parameters by 38.91% compared with the baseline while raising mean average precision by 1.0.

    Lu et al. · The Journal of Supercomputing, published online January 19, 2026 · DOI 10.1007/s11227-026-08234-w

  5. LitStreamDomain adaptation · Contrastive learning

    Source-free domain adaptation via multi-view contrastive learning

    The authors propose a three-phase method for adapting models without access to source data, combining a reliable sample memory, multi-view contrastive learning and noisy-label filtering, and report accuracy gains of about 2% over the second-best method on three benchmark datasets.

    Farhadi et al. · The Journal of Supercomputing, published online May 1, 2026 · DOI 10.1007/s11227-026-08460-2

See every The Journal of Supercomputing article on DeepDyve

Questions about The Journal of Supercomputing

How can I read The Journal of Supercomputing articles without a subscription?

10% of The Journal of Supercomputing articles published from January 1 to November 20, 2026 are open access and free to read. For subscription articles, Springer Nature lets authors post their accepted manuscript on their own website as soon as it’s accepted, and in a repository such as PubMed Central 12 months after publication. Your library may also subscribe. DeepDyve lets you read The Journal of Supercomputing articles online for $59 a month or $499 a year after a free first month.

What is the abbreviation for The Journal of Supercomputing?

The Journal of Supercomputing’s NLM (PubMed) abbreviation is J Supercomput.

What is the ISSN of The Journal of Supercomputing?

The print ISSN is 0920-8542 and the online ISSN is 1573-0484.

What is the impact factor of The Journal of Supercomputing?

The Journal of Supercomputing’s Journal Impact Factor is 3.5 (2025), and its 5-year Impact Factor is 3.1, according to Springer Nature’s published metrics.

Is The Journal of Supercomputing open access?

The Journal of Supercomputing is a hybrid journal: most articles are published under the subscription model, and authors can choose to publish open access. From January 1 to November 20, 2026, 82 of 804 The Journal of Supercomputing articles (10%) carry a Creative Commons license, per Crossref.

The Journal of Supercomputing articles on DeepDyve are licensed from Springer Nature. Journal facts as of 22 November 2026.

Start reading The Journal of Supercomputing today

One month free, then less than $2 a day. The Journal of Supercomputing, plus millions of articles from thousands of journals.

Start free month

Like this: