| Title | On Countermeasures Against the Thermal Covert Channel Attacks Targeting Many-core Systems |
| Publication Type | Conference Paper |
| Year of Publication | 2020 |
| Authors | Huang, H., Wang, X., Jiang, Y., Singh, A. K., Yang, M., Huang, L. |
| Conference Name | 2020 57th ACM/IEEE Design Automation Conference (DAC) |
| Keywords | Bit error rate, compositionality, covert channels, critical defense issue, Defense against Covert Channel Attack, DVFS technique, dynamic voltage frequency scaling technique, error statistics, Many-core systems, manycore systems, Multicore processing, multiprocessing systems, pubcrawl, Receivers, resilience, Resiliency, Scalability, security of data, signal detection, signal frequency scanning, TCC attack, Temperature sensors, Thermal Covert Channel, thermal covert channel attacks, thermal noise, three-step countermeasure, Transmitters |
| Abstract | Although it has been demonstrated in multiple studies that serious data leaks could occur to many-core systems thanks to the existence of the thermal covert channels (TCC), little has been done to produce effective countermeasures that are necessary to fight against such TCC attacks. In this paper, we propose a three-step countermeasure to address this critical defense issue. Specifically, the countermeasure includes detection based on signal frequency scanning, positioning affected cores, and blocking based on Dynamic Voltage Frequency Scaling (DVFS) technique. Our experiments have confirmed that on average 98% of the TCC attacks can be detected, and with the proposed defense, the bit error rate of a TCC attack can soar to 92%, literally shutting down the attack in practical terms. The performance penalty caused by the inclusion of the proposed countermeasures is only 3% for an 8x8 system. |
| DOI | 10.1109/DAC18072.2020.9218648 |
| Citation Key | huang_countermeasures_2020 |
On Countermeasures Against the Thermal Covert Channel Attacks Targeting Many-core Systems