Comparing Client- & Server-Side AEAD Encryption in Software-Defined Storage Systems
Bibliographic record
Abstract
To provide data confidentiality and establish data integrity with minimal performance overhead, “Authenticated Encryption with associated data” (AEAD) ciphers have become mandatory for implementing transport security in TLS 1.3. However, these ciphers have yet to find wide-range adoption within the domain of security at rest. This is problematic since software-defined storage (SDS) systems provide even smallscale organizations with a cost-effective method of storing large amounts of data. At present, AEAD ciphers at rest are most commonly used to encrypt objects in the object stores of AWS, Azure, and Google Cloud. However, these ciphers are not applied to client-side encryption for other storage formats, such as block or file storage, resulting in asymmetric security guarantees across storage services. On the server-side, AEAD encryption has, to the best of our knowledge, also not been widely adopted. Since neither AEAD encryption on the client- nor server-side has seen a wide-range adoption, this paper compares the benefits and downsides of employing AEAD encryption on the client- or server-side within SDS systems. To establish this comparison, we implemented a client-side and server-side encryption approach in the widely adopted Ceph SDS system. Our study demonstrates that incorporating AEAD ciphers can be achieved with a write performance loss of less than $5 \%$, while ensuring a higher security level than traditional encryption-at-rest approaches. Most importantly, we managed to achieve these security gains for data stored in all available storage formats in Ceph.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.006 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".