Using value locality to reduce memory encryption overhead in embedded processors

G. Keramidas, P. Petoumenos, A. Antonopoulos, S. Kaxiras, D. N. Serpanos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Memory encryption has gained much attention lately as a way to offer a secure environment to fight against software and hardware attacks. Many researchers provided memory encryption schemes whereby one or more levels of the memory hierarchy were encrypted using a cryptographic algorithm such as AES. Counter Mode (CM) encryption, also called One-Time-Pad (OTP) encryption, is proven to be quite effective for main memory encryption. However, CM encryption requires an extra sequence number (counter) to be associated with every memory location (12 block cacheline granularity is used). The perblock counters must be updated every time a block is written back to memory otherwise known-plaintext attacks may occur. Thus, the size of those counters is a critical parameter in the system design. In this work, we propose the use of silent stores as a method of providing the CM encryption with less overhead, Silent stores, i.e. stores, to memory that write the same value as already stored in that memory location, have been observed to occur frequently. These stores create redundant memory write-backs (and counter updates), so eliminating them will lower performance overheads introduced by the encyption/decryption process. Our initial results show significant benefits across the board indicating the promising nature of the proposed idea.

Original languageEnglish
Title of host publicationIEEE International Conference on Emerging Technologies and Factory Automation, ETFA
Pages632-637
Number of pages6
DOIs
Publication statusPublished - 1 Dec 2007
Externally publishedYes
Event12th IEEE International Conference on Emerging Technologies and Factory Automation, ETFA 2007 - Patras, Greece
Duration: 25 Sep 200728 Sep 2007

Other

Other12th IEEE International Conference on Emerging Technologies and Factory Automation, ETFA 2007
CountryGreece
CityPatras
Period25/9/0728/9/07

Fingerprint

Cryptography
Data storage equipment
Computer hardware
Systems analysis

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Keramidas, G., Petoumenos, P., Antonopoulos, A., Kaxiras, S., & Serpanos, D. N. (2007). Using value locality to reduce memory encryption overhead in embedded processors. In IEEE International Conference on Emerging Technologies and Factory Automation, ETFA (pp. 632-637). [4416828] https://doi.org/10.1109/EFTA.2007.4416828

Using value locality to reduce memory encryption overhead in embedded processors. / Keramidas, G.; Petoumenos, P.; Antonopoulos, A.; Kaxiras, S.; Serpanos, D. N.

IEEE International Conference on Emerging Technologies and Factory Automation, ETFA. 2007. p. 632-637 4416828.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Keramidas, G, Petoumenos, P, Antonopoulos, A, Kaxiras, S & Serpanos, DN 2007, Using value locality to reduce memory encryption overhead in embedded processors. in IEEE International Conference on Emerging Technologies and Factory Automation, ETFA., 4416828, pp. 632-637, 12th IEEE International Conference on Emerging Technologies and Factory Automation, ETFA 2007, Patras, Greece, 25/9/07. https://doi.org/10.1109/EFTA.2007.4416828
Keramidas G, Petoumenos P, Antonopoulos A, Kaxiras S, Serpanos DN. Using value locality to reduce memory encryption overhead in embedded processors. In IEEE International Conference on Emerging Technologies and Factory Automation, ETFA. 2007. p. 632-637. 4416828 https://doi.org/10.1109/EFTA.2007.4416828
Keramidas, G. ; Petoumenos, P. ; Antonopoulos, A. ; Kaxiras, S. ; Serpanos, D. N. / Using value locality to reduce memory encryption overhead in embedded processors. IEEE International Conference on Emerging Technologies and Factory Automation, ETFA. 2007. pp. 632-637
@inproceedings{ee778bb2cab840eeb69e452cffb25d1a,
title = "Using value locality to reduce memory encryption overhead in embedded processors",
abstract = "Memory encryption has gained much attention lately as a way to offer a secure environment to fight against software and hardware attacks. Many researchers provided memory encryption schemes whereby one or more levels of the memory hierarchy were encrypted using a cryptographic algorithm such as AES. Counter Mode (CM) encryption, also called One-Time-Pad (OTP) encryption, is proven to be quite effective for main memory encryption. However, CM encryption requires an extra sequence number (counter) to be associated with every memory location (12 block cacheline granularity is used). The perblock counters must be updated every time a block is written back to memory otherwise known-plaintext attacks may occur. Thus, the size of those counters is a critical parameter in the system design. In this work, we propose the use of silent stores as a method of providing the CM encryption with less overhead, Silent stores, i.e. stores, to memory that write the same value as already stored in that memory location, have been observed to occur frequently. These stores create redundant memory write-backs (and counter updates), so eliminating them will lower performance overheads introduced by the encyption/decryption process. Our initial results show significant benefits across the board indicating the promising nature of the proposed idea.",
author = "G. Keramidas and P. Petoumenos and A. Antonopoulos and S. Kaxiras and Serpanos, {D. N.}",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/EFTA.2007.4416828",
language = "English",
isbn = "1424408261",
pages = "632--637",
booktitle = "IEEE International Conference on Emerging Technologies and Factory Automation, ETFA",

}

TY - GEN

T1 - Using value locality to reduce memory encryption overhead in embedded processors

AU - Keramidas, G.

AU - Petoumenos, P.

AU - Antonopoulos, A.

AU - Kaxiras, S.

AU - Serpanos, D. N.

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Memory encryption has gained much attention lately as a way to offer a secure environment to fight against software and hardware attacks. Many researchers provided memory encryption schemes whereby one or more levels of the memory hierarchy were encrypted using a cryptographic algorithm such as AES. Counter Mode (CM) encryption, also called One-Time-Pad (OTP) encryption, is proven to be quite effective for main memory encryption. However, CM encryption requires an extra sequence number (counter) to be associated with every memory location (12 block cacheline granularity is used). The perblock counters must be updated every time a block is written back to memory otherwise known-plaintext attacks may occur. Thus, the size of those counters is a critical parameter in the system design. In this work, we propose the use of silent stores as a method of providing the CM encryption with less overhead, Silent stores, i.e. stores, to memory that write the same value as already stored in that memory location, have been observed to occur frequently. These stores create redundant memory write-backs (and counter updates), so eliminating them will lower performance overheads introduced by the encyption/decryption process. Our initial results show significant benefits across the board indicating the promising nature of the proposed idea.

AB - Memory encryption has gained much attention lately as a way to offer a secure environment to fight against software and hardware attacks. Many researchers provided memory encryption schemes whereby one or more levels of the memory hierarchy were encrypted using a cryptographic algorithm such as AES. Counter Mode (CM) encryption, also called One-Time-Pad (OTP) encryption, is proven to be quite effective for main memory encryption. However, CM encryption requires an extra sequence number (counter) to be associated with every memory location (12 block cacheline granularity is used). The perblock counters must be updated every time a block is written back to memory otherwise known-plaintext attacks may occur. Thus, the size of those counters is a critical parameter in the system design. In this work, we propose the use of silent stores as a method of providing the CM encryption with less overhead, Silent stores, i.e. stores, to memory that write the same value as already stored in that memory location, have been observed to occur frequently. These stores create redundant memory write-backs (and counter updates), so eliminating them will lower performance overheads introduced by the encyption/decryption process. Our initial results show significant benefits across the board indicating the promising nature of the proposed idea.

UR - http://www.scopus.com/inward/record.url?scp=47849105522&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47849105522&partnerID=8YFLogxK

U2 - 10.1109/EFTA.2007.4416828

DO - 10.1109/EFTA.2007.4416828

M3 - Conference contribution

AN - SCOPUS:47849105522

SN - 1424408261

SN - 9781424408269

SP - 632

EP - 637

BT - IEEE International Conference on Emerging Technologies and Factory Automation, ETFA

ER -