Answers are grounded in this paper and cite their sources.
Deleting a user’s data from a cloud service is rarely a single-store operation. Copies of the same data live in independently operated services, replicated stores, caches, indexes, logs, and backups, while erasure obligations, retention rules, and legal holds decide what must be removed, what must be kept, and by when. Following the PRISMA 2020 guideline where applicable, we report a systematic mapping study of 274 studies from a Semantic Scholar search (October 2026), one iteration of citation searching, and four directly added studies. A focused review covers 56 core studies that cross store, replica, or service boundaries or were identified in advance. We organize the field by six concerns — revocation versus erasure, discovery, ordering and integrity, verification and evidence, load and timeliness, and resurrection — and by the layer that enforces each technique. Of the mapped studies, 70% fall in clusters on erasure within a single store or outsourced data set; discovery, ordering, load, and resurrection are each coded for 9–19%, and few studies measure completeness or deadline adherence in production. We synthesize techniques and guarantees per concern, appraise the evidence, find no common correctness criterion for deletion across services, and derive an agenda for verifiable, timely, and non-resurrecting deletion. The map rests on one database and on cluster-level coding of non-core studies.
No takes yet. Share an insight, caveat, or question.
Roy et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: