Throughput and Power Efficiency Evaluation of Block Ciphers on Kepler and GCN GPUs Using Micro-Benchmark Analysis

Naoki NISHIKAWA; Keisuke IWAI; Hidema TANAKA; Takakazu KUROKAWA

doi:10.1587/transinf.E97.D.1506

Throughput and Power Efficiency Evaluation of Block Ciphers on Kepler and GCN GPUs Using Micro-Benchmark Analysis

Naoki NISHIKAWA, Keisuke IWAI, Hidema TANAKA, Takakazu KUROKAWA

Full Text Views

0

Share
Cite this

Summary :

Computer systems with GPUs are expected to become a strong methodology for high-speed encryption processing. Moreover, power consumption has remained a primary deterrent for such processing on devices of all sizes. However, GPU vendors are currently announcing their future roadmaps of GPU architecture development: Nvidia Corp. promotes the Kepler architecture and AMD Corp. emphasizes the GCN architecture. Therefore, we evaluated throughput and power efficiency of three 128-bit block ciphers on GPUs with recent Nvidia Kepler and AMD GCN architectures. From our experiments, whereas the throughput and per-watt throughput of AES-128 on Radeon HD 7970 (2048 cores) with GCN architecture are 205.0Gbps and 1.3Gbps/Watt respectively, those on Geforce GTX 680 (1536 cores) with Kepler architecture are, respectively, 63.9Gbps and 0.43Gbps/W; an approximately 3.2 times throughput difference occurs between AES-128 on the two GPUs. Next, we investigate the reasons for the throughput difference using our micro-benchmark suites. According to the results, we speculate that to ameliorate Kepler GPUs as co-processor of block ciphers, the arithmetic and logical instructions must be improved in terms of software and hardware.

Publication: IEICE TRANSACTIONS on Information Vol.E97-D No.6 pp.1506-1515

Publication Date: 2014/06/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E97.D.1506

Type of Manuscript: PAPER

Category: Fundamentals of Information Systems

Authors

Naoki NISHIKAWA
  National Defense Academy of Japan
Keisuke IWAI
  National Defense Academy of Japan
Hidema TANAKA
  National Defense Academy of Japan
Takakazu KUROKAWA
  National Defense Academy of Japan

Keyword

throughput, power efficiency, GPU, OpenCL, AES, Camellia, SC2000, Kepler, Graphics Core Next

Cite this

Copy

Naoki NISHIKAWA, Keisuke IWAI, Hidema TANAKA, Takakazu KUROKAWA, "Throughput and Power Efficiency Evaluation of Block Ciphers on Kepler and GCN GPUs Using Micro-Benchmark Analysis" in IEICE TRANSACTIONS on Information, vol. E97-D, no. 6, pp. 1506-1515, June 2014, doi: 10.1587/transinf.E97.D.1506.
Abstract: Computer systems with GPUs are expected to become a strong methodology for high-speed encryption processing. Moreover, power consumption has remained a primary deterrent for such processing on devices of all sizes. However, GPU vendors are currently announcing their future roadmaps of GPU architecture development: Nvidia Corp. promotes the Kepler architecture and AMD Corp. emphasizes the GCN architecture. Therefore, we evaluated throughput and power efficiency of three 128-bit block ciphers on GPUs with recent Nvidia Kepler and AMD GCN architectures. From our experiments, whereas the throughput and per-watt throughput of AES-128 on Radeon HD 7970 (2048 cores) with GCN architecture are 205.0Gbps and 1.3Gbps/Watt respectively, those on Geforce GTX 680 (1536 cores) with Kepler architecture are, respectively, 63.9Gbps and 0.43Gbps/W; an approximately 3.2 times throughput difference occurs between AES-128 on the two GPUs. Next, we investigate the reasons for the throughput difference using our micro-benchmark suites. According to the results, we speculate that to ameliorate Kepler GPUs as co-processor of block ciphers, the arithmetic and logical instructions must be improved in terms of software and hardware.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.E97.D.1506/_p

Copy

@ARTICLE{e97-d_6_1506,
author={Naoki NISHIKAWA, Keisuke IWAI, Hidema TANAKA, Takakazu KUROKAWA, },
journal={IEICE TRANSACTIONS on Information},
title={Throughput and Power Efficiency Evaluation of Block Ciphers on Kepler and GCN GPUs Using Micro-Benchmark Analysis},
year={2014},
volume={E97-D},
number={6},
pages={1506-1515},
abstract={Computer systems with GPUs are expected to become a strong methodology for high-speed encryption processing. Moreover, power consumption has remained a primary deterrent for such processing on devices of all sizes. However, GPU vendors are currently announcing their future roadmaps of GPU architecture development: Nvidia Corp. promotes the Kepler architecture and AMD Corp. emphasizes the GCN architecture. Therefore, we evaluated throughput and power efficiency of three 128-bit block ciphers on GPUs with recent Nvidia Kepler and AMD GCN architectures. From our experiments, whereas the throughput and per-watt throughput of AES-128 on Radeon HD 7970 (2048 cores) with GCN architecture are 205.0Gbps and 1.3Gbps/Watt respectively, those on Geforce GTX 680 (1536 cores) with Kepler architecture are, respectively, 63.9Gbps and 0.43Gbps/W; an approximately 3.2 times throughput difference occurs between AES-128 on the two GPUs. Next, we investigate the reasons for the throughput difference using our micro-benchmark suites. According to the results, we speculate that to ameliorate Kepler GPUs as co-processor of block ciphers, the arithmetic and logical instructions must be improved in terms of software and hardware.},
keywords={},
doi={10.1587/transinf.E97.D.1506},
ISSN={1745-1361},
month={June},}

Copy

TY - JOUR
TI - Throughput and Power Efficiency Evaluation of Block Ciphers on Kepler and GCN GPUs Using Micro-Benchmark Analysis
T2 - IEICE TRANSACTIONS on Information
SP - 1506
EP - 1515
AU - Naoki NISHIKAWA
AU - Keisuke IWAI
AU - Hidema TANAKA
AU - Takakazu KUROKAWA
PY - 2014
DO - 10.1587/transinf.E97.D.1506
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E97-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2014
AB - Computer systems with GPUs are expected to become a strong methodology for high-speed encryption processing. Moreover, power consumption has remained a primary deterrent for such processing on devices of all sizes. However, GPU vendors are currently announcing their future roadmaps of GPU architecture development: Nvidia Corp. promotes the Kepler architecture and AMD Corp. emphasizes the GCN architecture. Therefore, we evaluated throughput and power efficiency of three 128-bit block ciphers on GPUs with recent Nvidia Kepler and AMD GCN architectures. From our experiments, whereas the throughput and per-watt throughput of AES-128 on Radeon HD 7970 (2048 cores) with GCN architecture are 205.0Gbps and 1.3Gbps/Watt respectively, those on Geforce GTX 680 (1536 cores) with Kepler architecture are, respectively, 63.9Gbps and 0.43Gbps/W; an approximately 3.2 times throughput difference occurs between AES-128 on the two GPUs. Next, we investigate the reasons for the throughput difference using our micro-benchmark suites. According to the results, we speculate that to ameliorate Kepler GPUs as co-processor of block ciphers, the arithmetic and logical instructions must be improved in terms of software and hardware.
ER -