Run this audit before uploading to the FAST HotCRP site. FAST — the USENIX Conference on File and
Storage Technologies — publishes open access via USENIX, reviews double-blind, and runs
two deadlines a year. The first decision you make is not about the PDF but about which
deadline you are aiming at. Every number below was read from the FAST '26/'27 calls on 2026-07-09
via search renderings of the usenix.org URLs (see resources/official-source-map.md); treat them
as a one-cycle snapshot and reopen the live call first.
FAST '27 runs a Spring deadline (17 Mar 2026) and a Fall deadline (15 Sep 2026), each with its own HotCRP instance, author-response window, and notification date. Consequences:
FAST expects title, authors, abstract, and topics registered in HotCRP; some cycles require this a few days before the full-paper cutoff. Register with the real title and abstract — they drive reviewer bidding and PC match — not a placeholder you intend to swap.
acmart, not
IEEE 10+2. Do not carry a two-column ACM/IEEE habit over.FAST review is double-blind: anonymize the PDF and references and refer to your own prior work in the third person. Storage papers leak identity through devices, tools, and traces, so the surface is wide:
# Mechanical pass on the submission PDF and any anonymized artifact archive
pdfinfo paper.pdf | grep -Ei 'author|creator|producer'
pdftotext paper.pdf - | grep -nEi 'github\.com/[a-z0-9-]+|zenodo\.org/record|our (system|lab|group)|acknowledg|grant' | head
unzip -l artifact.zip | grep -Ei '\.git/|\.DS_Store|/home/|/Users/|hostname' | head
grep -rni 'university\|@[a-z0-9.]*\.edu\|internal cluster\|<companyname>' artifact/ --include='*.md' | head
The storage-specific leaks: a system named after your lab or product, a cluster/hostname in a config or SMART dump, a trace hosted on a personal or institutional URL, a datacenter name in a field study, and commit metadata inside a zipped repository. Re-host artifacts and traces behind an anonymizing location before upload.
USENIX values sharing. Concretely for the submission:
| Finding at audit time | Severity | Real fix |
|---|---|---|
| Body over the 12/6-page limit | Desk-reject-grade | Cut or move to the artifact; references don't absorb body text |
| Wrong template (ACM/IEEE, altered margins/font) | Named format-reject ground | Recompile in the USENIX template; recover space editorially |
| Identity leak in PDF, artifact, tool, or trace URL | Anonymity violation | Re-anonymize and re-host; scrub PDF and SMART/config metadata |
| Registered on the wrong deadline's HotCRP | No valid submission there | Re-register on the correct Spring/Fall instance before its cutoff |
| No device/firmware detail in the evaluation | Storage-soundness weakness | Add the testbed table before submission, not in rebuttal |
| Same work under review elsewhere | Concurrent-submission violation | Withdraw one venue; a one-shot revision also blocks parallel submission |
[FAST submission status] ready / blocked / needs work
[Target deadline] Spring / Fall <date + correct HotCRP URL>
[Registration] title/abstract/authors/topics/conflicts locked? yes/no
[Format] pages used (body/refs), USENIX template compliance
[Anonymity] clean / leaks: <where>
[Availability] artifact/trace bundle present? device+firmware provenance stated?
[Fix queue] <ordered, with owners and dates before the AoE cutoff>