OSDI '27 Call for Papers | USENIX

Sponsored by USENIX in cooperation with ACM SIGOPS
The 21st USENIX Symposium on Operating Systems Design and Implementation (OSDI '27) will take place on July 7–9, 2027, at Hyatt Regency Baltimore Inner Harbor in Baltimore, MD, USA.
OSDI, the USENIX Symposium on Operating Systems Design and Implementation, brings together professionals from academic and industrial backgrounds in a premier forum for discussing the design, implementation, and implications of systems software. OSDI emphasizes innovative research and quantified or insightful experiences in systems design and implementation.
OSDI takes a broad view of the systems area and solicits contributions from many fields of systems practice, including operating systems, file and storage systems, distributed systems, cloud computing, mobile systems, secure and reliable systems, formal verification of systems software, systems aspects of big data, systems for machine learning, embedded systems, virtualization, networking as it relates to operating systems, and management and troubleshooting of complex systems. We also welcome work that explores the interface between systems and related areas such as artificial intelligence, computer architecture, networking, programming languages, analytics, and databases. We encourage contributions with highly original ideas, new approaches, and groundbreaking results.
Compared to previous years, OSDI '27:
Continued Important Policies from 2026:
Submissions will be judged on novelty, significance, interest, clarity, relevance, and correctness. An excellent paper will:
Continuing in OSDI's tradition, the OSDI program committee will select papers that offer significant contributions to computer systems research and align with the interests of our community. Reviewers will evaluate submissions based on topic relevance to computer systems and potential to impact future research and practices in computer systems. Additionally, submissions must offer insights that intrigue a substantial portion of OSDI attendees. Papers with little overlap with the program committee's interests are less likely to be accepted. Papers applying systems techniques to problems in new domains (e.g., ML) should engage with the relevant systems prior work, baseline against those known techniques, and spell out the new systems challenges the setting creates.
OSDI '27 will allow an author to be listed on at most eight submissions. If an author is listed on more than eight submissions to the conference, the program co-chairs will desk-reject surplus submissions.
The program committee will not aim on a particular acceptance rate, but instead try to accept a number of strong papers that can fit reasonably within a 3-day program. At the same time, the committee realizes that very low acceptance rates harm conferences because they detrimentally bias the proceedings towards flawless presentations of narrower ideas or results.
Currently, submission rates are in great flux, and we will adapt the review process as needed. OSDI '27 may include an early review process in which program committee members only examine the first few pages of papers. Whether OSDI '27 includes this stage in the process will depend on the expected number of submissions and the experiences of conferences that are trying this process. Whether this process occurs and the subsequent details will be added to the Call for Papers at least one month before the paper submission deadline.
Update of October 2026: OSDI '27 will use an early review round if submission volume necessitates it. In that case, reviewers will read at least the first two pages of each paper and vote on whether it advances to full review. They will consider whether a paper is in scope for OSDI, is understandable, and has a reasonable chance of making a significant contribution to systems research. Those who vote "no" will provide feedback explaining why.
Each submission must also name one of its authors as a Volunteer Candidate Senior Reviewer who is not already on the OSDI '27 Program Committee. This person must have a PhD or equivalent research experience, and a significant publication record at OSDI, SOSP, or comparable systems venues (e.g., NSDI, EuroSys, ASPLOS, ATC, FAST). If additional reviewers are needed, the program co-chairs will evaluate these candidates and invite selected individuals to review up to four papers each in the first full review round. Those who review will be acknowledged as external reviewers in the front matter of the proceedings.
A paper is exempt if at least one author is on the OSDI '27 program committee, if no author meets these criteria, or if other exceptional circumstances apply (cleared by the program co-chairs before the deadline). Exempt papers simply indicate this on the submission form. Exemption status is visible only to the program co-chairs and has no effect on how a paper is reviewed. We welcome submissions from teams new to OSDI.
Papers accompanied by nondisclosure agreement forms will not be considered. All submissions will be treated as confidential by USENIX and by USENIX reviewers prior to publication; rejected submissions will be permanently treated as confidential.
Simultaneous submission of the same work to multiple venues, submission of previously published work, and plagiarism constitute dishonesty or fraud. Submission of a paper whose text was wholly or largely generated by AI also constitutes fraud. Using AI editing tools is acceptable. USENIX, like other scientific and technical conferences and journals, prohibits these practices and may, on the recommendation of a program chair, take action against authors who have committed them. See the USENIX Conference Submissions Policy for details.
A system described by an OSDI submission may be developed with the assistance of coding agents, but a paper must clearly disclose their use (e.g., in the "Implementation" section). A key expectation of any OSDI paper is that it "draw appropriate conclusions." Thus,