Pith Number

pith:Z73BFPAM

pith:2026:Z73BFPAMADSAXZRBULAO2JXGUJ

not attested not anchored not stored refs resolved

HE-PIM: Demystifying Homomorphic Operations on a Real-world Processing-in-Memory System

Antonio J. Pe\~na, Harshita Gupta, Jaewoo Park, Juan G\'omez-Luna, Konstantinos Kanellopoulos, Mayank Kabra, Mohammad Sadrosadati, Nisa Bostanc{\i}, Onur Mutlu, Phillip Widdowson, Priyam Mehta, Tathagata Barik

Processing-in-memory systems become competitive with CPUs and GPUs for homomorphic encryption when equipped with native modular multiplication and efficient data movement.

arxiv:2605.12841 v1 · 2026-05-13 · cs.CR

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{Z73BFPAMADSAXZRBULAO2JXGUJ}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Despite these limits, PIM can be a viable alternative to state-of-the-art CPU and GPU systems for HE when equipped with native modular multiplication and efficient inter-PIM data movement.

C2weakest assumption

The UPMEM PIM system and the chosen set of HE kernels are representative of future general-purpose PIM hardware and of the workloads that will actually be deployed in encrypted databases and machine learning.

C3one line summary

Characterization of HE kernels on commercial UPMEM PIM identifies modular multiplication and per-bank capacity as dominant bottlenecks and concludes PIM becomes competitive with CPU/GPU once those are addressed.

References

300 extracted · 300 resolved · 3 Pith anchors

[1] On Data Banks and Privacy Homo- morphisms, 1978

[2] On Lattices, Learning with Errors, Random Linear Codes, and Cryptog- raphy, 2009

[3] On Ideal Lattices and Learning with Errors over Rings, 2010

[4] Fully Homomorphic Encryption Using Ideal Lattices, 2009

[5] A Fully Homomorphic Encryption Scheme, 2009

Cited by

1 paper in Pith

Taking Cryptography Out of the Data Path via Near-Memory Processing in DRAM

Receipt and verification

First computed	2026-05-18T03:09:11.963154Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

cff612bc0c00e40be621a2c0ed26e6a2714213e8ba072f4e30b2f69d92a0bb4c

Aliases

arxiv: 2605.12841 · arxiv_version: 2605.12841v1 · doi: 10.48550/arxiv.2605.12841 · pith_short_12: Z73BFPAMADSA · pith_short_16: Z73BFPAMADSAXZRB · pith_short_8: Z73BFPAM

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/Z73BFPAMADSAXZRBULAO2JXGUJ \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: cff612bc0c00e40be621a2c0ed26e6a2714213e8ba072f4e30b2f69d92a0bb4c

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "5c8e08165de876f8790371e8b182a6f8d568a488b3c7a195d45be7358afab025",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CR",
    "submitted_at": "2026-05-13T00:36:03Z",
    "title_canon_sha256": "a0ef9b58a4441a89242c0711faeb5bb95cfaba22dca1b59bece8f05fba483077"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.12841",
    "kind": "arxiv",
    "version": 1
  }
}