SIGCOMM '27 Call For Papers
SIGCOMM 2027 will see several major changes including (see below for details):
- Two kinds of paper acceptances: Some papers will be accepted to appear at the SIGCOMM 2027 conference and some will be accepted to appear as a research article in SIGCOMM’s Computer Communications Review (CCR). Authors can opt out of this latter acceptance category if they wish to resubmit their paper elsewhere.
- Two kinds of in-person presentations for papers accepted to appear in the conference: Some papers will be presented in the conference plenary sessions, and all papers will be presented in small-scale discussion sessions.
- A new approach to reviewing: The SIGCOMM program committee will use a “collaborative reviewing” approach where reviewers are in discussion with each other from the start of the reviewing process.
- A general accessibility requirement: The motivation and technical content of a submitted paper should be understandable by most researchers in the general area of networking and networked systems. Accessibility will be a criteria in the reviewing process.
- A revised focus for the experience track: Submissions to the experience track should present insights found in the course of deploying and/or operating networking technologies and techniques, and will be judged on the depth of those insights.
- No rebuttals or one-shot revisions
What has not changed: the scope of the conference.
The ACM SIGCOMM 2027 conference seeks papers describing significant research contributions and/or significant deployment experiences in networking and networked systems. SIGCOMM takes a broad view of these topics, which includes (but is not limited to):
- All types of computer networks, including mobile, wide-area, datacenter, home, and enterprise networks.
- All types of wired and wireless networking technologies, including optics, radio, acoustic, and visible light-based communication.
- All aspects of networks and networked systems, such as network architecture, packet-switched and circuit-switched hardware and software, virtualization, mobility, resource management, performance, energy consumption and environmental impact, topology, robustness, security, diagnosis, verification, privacy, economics, evolution, and interactions with applications.
- All parts of the network life cycle, including planning, designing, building, deploying, operating, troubleshooting, migrations, and end-of-life.
- All approaches and techniques applied to networking, including theory, analysis, experimentation, and AI/machine learning.
SIGCOMM 2027 will introduce several changes to our past SIGCOMM conference practices. These changes address how papers are accepted, presented, reviewed, and written.
Decisions on submissions: For each paper, the program committee (PC) can reach one of three decisions:
- Accepted into the SIGCOMM conference: These papers are accepted into the SIGCOMM conference. Each of the accepted papers will be assigned a shepherd to provide advice on revisions.
- Accepted into CCR: These papers are rejected from the SIGCOMM conference but accepted into CCR to appear as regular CCR research articles. Authors can decline the offer to appear in CCR and are free to resubmit their paper elsewhere.
- Rejected: These papers are rejected from both the SIGCOMM conference and CCR.
Presentations:
- There will be three different forms of paper presentations:
- In-person presentations in single-track plenary sessions.
- In-person small-scale discussion sessions similar to what is described in this article about ConversTations. More information about these sessions will be available at a later date.
- Video presentations for all accepted papers, which will be made available online before the conference.
- All papers accepted into the SIGCOMM conference must provide a video presenting the work, and must have an author present the paper at an in-person discussion session.
- A relatively small subset of papers accepted into the conference will be presented in single-track plenary sessions. We will assign a “talk shepherd” for each of these presentations, to help in their preparation. To encourage active discussion at the conference, each plenary talk will be followed by a respondent who will initiate a conversation about the paper with a few remarks. The papers presented in the plenary sessions will be chosen by the program committee based on which paper presentations they think will best trigger discussion and inspire future work. These papers should not be seen as “better” than the papers not chosen for plenary presentations, as all papers accepted to the SIGCOMM conference are full-fledged SIGCOMM papers, selected by the PC according to the traditional high standards of creativity, technical depth, and exposition. We discourage authors from indicating these plenary presentations on their publication records, as it is in no way a differentiating measure of quality.
- As in the past two years, SIGCOMM 2027 will have remote “pods” as described here. Some of these pods may want to arrange Q/A sessions with authors during the conference, and we strongly urge authors to make themselves available for these sessions.
Reviewing:
- From the very start of the reviewing process, the reviewers of a paper will be in conversation, helping each other understand and appreciate the paper’s contributions. This “collaborative reviewing” process makes it much easier for a reviewer to (i) ask questions while reading the paper, so their review can be more fully informed and (ii) gently express differences of opinion with other reviewers before any reviewers have committed to an accept/reject decision. After the PC has reached a decision on the paper, the set of reviewers produce a single but relatively short review that provides the authors with the consensus opinion of the paper. Authors should be aware that the PC feedback will be more limited than in the past, much more like a PC summary than a set of detailed individual (and sometimes contradictory) reviews.
- In contrast to some previous years, there will be no rebuttals or one-shot revisions.
- Each submitted paper must, in the submission form, provide the name of an author willing to review one or more papers from the conference. These “reciprocal” reviewers will not be part of the official PC but will provide external input into the reviewing process. Given the ever-increasing number of submissions, these reciprocal reviewers are necessary to increase the size and diversity of the reviewer pool. A reciprocal reviewer may be asked to review as many papers as they have submitted. Except in exceptional circumstances, reciprocal reviewers must hold a PhD.
General accessibility:
- The SIGCOMM conference is an opportunity for the SIGCOMM community to come together to discuss a new set of research contributions. Unfortunately, when papers are written and presented for other experts in the field, there is a tendency for attendees to only read, and/or attend presentations of, papers in their own field. To create a broader sharing of ideas, each paper should be written and presented so that the core material can be understood by most researchers working in the area of networks and networked systems. Moreover, the paper should motivate, in generally accessible language, why we as a community should care about the problem or topic being addressed. The PC will reject papers whose presentation of the motivation or approach will likely not be understood by most SIGCOMM attendees. Of course, some details of a paper’s techniques may be hard for inexpert reviewers to fully grasp, but the basic gist of the approach should be accessible.
A revised experience track: SIGCOMM 2027 will continue to have two submission categories, research and experience, with survey and tutorial papers remaining out of scope. Papers in both tracks represent valuable contributions to our understanding, but they contribute in somewhat different ways.
- Research track submissions should significantly advance the state of the art in areas within the conference’s purview by, for instance, proposing and developing novel ideas, rigorously evaluating or re-evaluating existing ideas, or increasing our understanding of current usage patterns and properties. This is the traditional expectation for research track papers.
- Experience track submissions should present insights found in the course of deploying and/or operating designs within the conference’s purview. What is new this year is that the emphasis should be on the lessons learned, and papers will be evaluated on the depth of insight in these lessons rather than on the size or duration of the deployment. The lessons learned must be clearly explained and supported by observations from the deployment.
At paper registration time, authors must explicitly register in the submission form whether their paper is to be considered for the research or experience track. Authors should also identify experience track papers by prepending the title with “Experience:”
Submissions should be in two-column, 10-point format, and can be up to 12 pages in length with as many additional pages as necessary for references and optional appendices. Reviewers are not required to read appendices or consider them in their review. Authors should thus ensure that their paper is complete and self-contained without the appendix. For example, if the appendix provides details of a proof or experiment, the body should summarize the key result. Appendices may also include non-traditional material, such as videos, datasets, and code, all appropriately anonymized.
Please see the Submission Instructions for details on paper formatting, registration, anonymity, and other important issues relevant to your submission. Please note that we require authors to include their ORCID at submission time. If you are unable to generate an ORCID or wish to submit your paper anonymously, please contact the chairs.
Accepted papers may be shepherded by a member of the PC to ensure reviewer feedback is appropriately addressed. The shepherd will also review appendices and must approve their necessity.
For accepted papers, the official publication date is July 18th, 2027, which is three weeks before the first day of the conference. This may affect the deadline for any patent filings related to published work.
Submission site: https://sigcomm27.hotcrp.com
Each author is limited to at most eight submissions across both tracks. If an author is listed on more than eight submissions to the conference, the program co-chairs will desk-reject surplus submissions. Additionally, we will enforce this policy when accounting for requested authorship changes to accepted papers.
All papers must include, on the submission form, a description of any ethical issues raised by the work. The PC takes a broad view of what constitutes an ethical concern, and the PC chairs may reach out to authors during the review process if questions arise.
Authors should consult the ACM Policy on Authorship when it comes to the use of GenAI. Authors will be required to state their compliance with this policy on the submission form.
The authors of papers accepted into the conference will be invited to submit supporting materials to the Artifact Evaluation process. Artifact Evaluation is run by a separate committee that will assess how well the submitted artifacts support the work described in the accepted papers. Participation in Artifact Evaluation is voluntary and has no influence on paper acceptance, but it is strongly encouraged. Papers that go through the Artifact Evaluation process successfully will receive badges printed on the papers themselves. Additional details on the Artifact Evaluation process can be found on the SIGCOMM 2027 website.