Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions

Kotaro Inoue; Takuya Suzuki; Hayato Yamana

doi:10.1145/3560827.3563374

Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions

Kotaro Inoue, Takuya Suzuki, Hayato Yamana

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

More and more data analysis is being outsourced due to the spread of cloud computing. Therefore, protection of the data from privacy violations and information leaks is required. In particular, homomorphic encryption, which allows computation to be performed with encrypted data, is being actively studied as one of the protection method. Ring learning with errors based homomorphic encryption schemes support packing which allows to pack several elements into slots of a plaintext and ciphertext. A trace-type function, which combines shifting slots (rotation) and homomorphic addition to obtain summation of slots, is often used in homomorphic encryption applications and acceleration of the trace-type function is important. In this paper, we further accelerate the trace-type function using Intel AVX512 compared to existing optimized trace-type function with loop unrolling. The results show that our AVX512 version was 1.05-2.30 times speedup compared to the non-AVX512 version.

Original language	English
Title of host publication	WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022
Publisher	Association for Computing Machinery, Inc
Pages	47-52
Number of pages	6
ISBN (Electronic)	9781450398770
DOIs	https://doi.org/10.1145/3560827.3563374
Publication status	Published - 2022 Nov 7
Event	10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, WAHC 2022 - Co-located with CCS 2022 - Los Angeles, United States Duration: 2022 Nov 7 → …

Publication series

Name	WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022

Conference

Conference	10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, WAHC 2022 - Co-located with CCS 2022
Country/Territory	United States
City	Los Angeles
Period	22/11/7 → …

Keywords

homomorphic encryption
secure computation
simd

ASJC Scopus subject areas

Computer Networks and Communications
Software

Access to Document

10.1145/3560827.3563374

Cite this

Inoue, K., Suzuki, T., & Yamana, H. (2022). Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions. In WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022 (pp. 47-52). (WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022). Association for Computing Machinery, Inc. https://doi.org/10.1145/3560827.3563374

Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions. / Inoue, Kotaro; Suzuki, Takuya ; Yamana, Hayato.
WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022. Association for Computing Machinery, Inc, 2022. p. 47-52 (WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Inoue, K, Suzuki, T & Yamana, H 2022, Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions. in WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022. WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022, Association for Computing Machinery, Inc, pp. 47-52, 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, WAHC 2022 - Co-located with CCS 2022, Los Angeles, United States, 22/11/7. https://doi.org/10.1145/3560827.3563374

Inoue K, Suzuki T , Yamana H. Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions. In WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022. Association for Computing Machinery, Inc. 2022. p. 47-52. (WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022). doi: 10.1145/3560827.3563374

Inoue, Kotaro ; Suzuki, Takuya ; Yamana, Hayato. / Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions. WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022. Association for Computing Machinery, Inc, 2022. pp. 47-52 (WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022).

@inproceedings{930ea7c46173484a8f159c7bb5926a25,

title = "Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions",

abstract = "More and more data analysis is being outsourced due to the spread of cloud computing. Therefore, protection of the data from privacy violations and information leaks is required. In particular, homomorphic encryption, which allows computation to be performed with encrypted data, is being actively studied as one of the protection method. Ring learning with errors based homomorphic encryption schemes support packing which allows to pack several elements into slots of a plaintext and ciphertext. A trace-type function, which combines shifting slots (rotation) and homomorphic addition to obtain summation of slots, is often used in homomorphic encryption applications and acceleration of the trace-type function is important. In this paper, we further accelerate the trace-type function using Intel AVX512 compared to existing optimized trace-type function with loop unrolling. The results show that our AVX512 version was 1.05-2.30 times speedup compared to the non-AVX512 version.",

keywords = "homomorphic encryption, secure computation, simd",

author = "Kotaro Inoue and Takuya Suzuki and Hayato Yamana",

note = "Publisher Copyright: {\textcopyright} 2022 ACM.; 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, WAHC 2022 - Co-located with CCS 2022 ; Conference date: 07-11-2022",

year = "2022",

month = nov,

day = "7",

doi = "10.1145/3560827.3563374",

language = "English",

series = "WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022",

publisher = "Association for Computing Machinery, Inc",

pages = "47--52",

booktitle = "WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022",

}

TY - GEN

T1 - Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions

AU - Inoue, Kotaro

AU - Suzuki, Takuya

AU - Yamana, Hayato

PY - 2022/11/7

Y1 - 2022/11/7

N2 - More and more data analysis is being outsourced due to the spread of cloud computing. Therefore, protection of the data from privacy violations and information leaks is required. In particular, homomorphic encryption, which allows computation to be performed with encrypted data, is being actively studied as one of the protection method. Ring learning with errors based homomorphic encryption schemes support packing which allows to pack several elements into slots of a plaintext and ciphertext. A trace-type function, which combines shifting slots (rotation) and homomorphic addition to obtain summation of slots, is often used in homomorphic encryption applications and acceleration of the trace-type function is important. In this paper, we further accelerate the trace-type function using Intel AVX512 compared to existing optimized trace-type function with loop unrolling. The results show that our AVX512 version was 1.05-2.30 times speedup compared to the non-AVX512 version.

AB - More and more data analysis is being outsourced due to the spread of cloud computing. Therefore, protection of the data from privacy violations and information leaks is required. In particular, homomorphic encryption, which allows computation to be performed with encrypted data, is being actively studied as one of the protection method. Ring learning with errors based homomorphic encryption schemes support packing which allows to pack several elements into slots of a plaintext and ciphertext. A trace-type function, which combines shifting slots (rotation) and homomorphic addition to obtain summation of slots, is often used in homomorphic encryption applications and acceleration of the trace-type function is important. In this paper, we further accelerate the trace-type function using Intel AVX512 compared to existing optimized trace-type function with loop unrolling. The results show that our AVX512 version was 1.05-2.30 times speedup compared to the non-AVX512 version.

KW - homomorphic encryption

KW - secure computation

KW - simd

UR - http://www.scopus.com/inward/record.url?scp=85142628739&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85142628739&partnerID=8YFLogxK

U2 - 10.1145/3560827.3563374

DO - 10.1145/3560827.3563374

M3 - Conference contribution

AN - SCOPUS:85142628739

T3 - WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022

SP - 47

EP - 52

BT - WAHC 2022 - Proceedings of the 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, co-located with CCS 2022

PB - Association for Computing Machinery, Inc

T2 - 10th Workshop on Encrypted Computing and Applied Homomorphic Cryptography, WAHC 2022 - Co-located with CCS 2022

Y2 - 7 November 2022

ER -

Acceleration of Homomorphic Unrolled Trace-Type Function using AVX512 instructions

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this