Search

Randomized, skewed caches (RSCs) such as CEASER-S have recently received much attention to defend against contention-based cache side channels. By randomizing and regularly changing the mapping(s) of addresses to cache sets, these techniques are designed to obfuscate the leakage of memory access patterns. However, new attack techniques, e.g., Prime+Prune+Probe, soon demonstrated the limits of RSCs as they allow attackers to more quickly learn which addresses contend in the cache and use this information to circumvent the randomization. To yet maintain side-channel resilience, RSCs must change the random mapping(s) more frequently with adverse effects on performance and implementation complexity. This work aims to make randomization-based approaches more robust to allow for reduced re-keying rates and presents Chameleon Cache. Chameleon Cache extends RSCs with a victim cache (VC) to decouple contention in the RSC from evictions observed by the user. The VC allows Chameleon Cache to make additional use of the multiple mappings RSCs provide to translate addresses to cache set indices: when a cache line is evicted from the RSC to the VC under one of its mappings, the VC automatically reinserts this evicted line back into the RSC by using a different mapping. As a result, the effects of previous RSC set contention are hidden and Chameleon Cache exhibits side-channel resistance and eviction patterns similar to fully associative caches with random replacement. We show that Chameleon Cache has performance overheads of < 1% and stress that VCs are more generically helpful to increase side-channel resistance and re-keying intervals of randomized caches., Comment: 12 pages, 9 figures, 6 algorithms, 1 table

Over the last two decades, the danger of sharing resources between programs has been repeatedly highlighted. Multiple side-channel attacks, which seek to exploit shared components for leaking information, have been devised, mostly targeting shared caching components. In response, the research community has proposed multiple cache designs that aim at curbing the source of side channels. With multiple competing designs, there is a need for assessing the level of security against side-channel attacks that each design offers. In this work we propose CacheFX, a flexible framework for assessing and evaluating the resilience of cache designs to side-channel attacks. CacheFX allows the evaluator to implement various cache designs, victims, and attackers, as well as to exercise them for assessing the leakage of information via the cache. To demonstrate the power of CacheFX, we implement multiple cache designs and replacement algorithms, and devise three evaluation metrics that measure different aspects of the caches:(1) the entropy induced by a memory access; (2) the complexity of building an eviction set; and (3) protection against cryptographic attacks. Our experiments highlight that different security metrics give different insights to designs, making a comprehensive analysis mandatory. For instance, while eviction-set building was fastest for randomized skewed caches, these caches featured lower eviction entropy and higher practical attack complexity. Our experiments show that all non-partitioned designs allow for effective cryptographic attacks. However, in state-of-the-art secure caches, eviction-based attacks are more difficult to mount than occupancy-based attacks, highlighting the need to consider the latter in cache design.

Yuan Li,1– 3,&ast; Yaxin Zhou,4,5,&ast; Miao Chen,6 Mei R Fu,7 Biru Luo,1– 3 Pengming Yu,4,5,8 Hong Zheng,1,9 Fangfei Liu1,9 1Nursing Department, West China Second University Hospital, Sichuan University, Chengdu, People’s Republic of China; 2Key Laboratory of Birth Defects and Related Diseases of Women and Children (Sichuan University), Ministry of Education, Chengdu, People’s Republic of China; 3West China School of Nursing, Sichuan University, Chengdu, People’s Republic of China; 4Rehabilitation Medicine Center, Sichuan University, Chengdu, People’s Republic of China; 5West China Hospital, Sichuan University, Chengdu, People’s Republic of China; 6Department of Cardiovascular Surgery, West China Hospital, Sichuan University, Chengdu, People’s Republic of China; 7Rutgers University, School of Nursing, Camden, NJ, USA; 8Key Laboratory of Rehabilitation Medicine in Sichuan Province, Chengdu, People’s Republic of China; 9Department of Pediatric Cardiology, West China Second University Hospital, Sichuan University, Chengdu, People’s Republic of China&ast;These authors contributed equally to this workCorrespondence: Biru Luo, Nursing Department, West China Second University Hospital, Sichuan University, No. 20, Section 3, South Renmin Road, Chengdu, Sichuan, 610041, People’s Republic of China, Tel +86-1 818 060 9180, Email luomr@scu.edu.cn Miao Chen, Department of Cardiovascular Surgery, West China Hospital, Sichuan University, No. 37, Guoxue Alley, Wuhou District, Chengdu, Sichuan, 610041, People’s Republic of China, Tel +86-1 898 060 6107, Email 540129343@qq.comAbstract: Progress in medical and surgical care has tremendously improved the survival rates of children with congenital heart disease (CHD). However, reduced aerobic capacity and health-related issues remain a threaten to quality survival and prevention of related complications among children and adolescents with CHD. This research program aims to develop and evaluate a WeCh

Recent security vulnerabilities that target speculative execution (e.g., Spectre) present a significant challenge for processor design. The highly publicized vulnerability uses speculative execution to learn victim secrets by changing cache state. As a result, recent computer architecture research has focused on invisible speculation mechanisms that attempt to block changes in cache state due to speculative execution. Prior work has shown significant success in preventing Spectre and other vulnerabilities at modest performance costs. In this paper, we introduce speculative interference attacks, which show that prior invisible speculation mechanisms do not fully block these speculation-based attacks. We make two key observations. First, misspeculated younger instructions can change the timing of older, bound-to-retire instructions, including memory operations. Second, changing the timing of a memory operation can change the order of that memory operation relative to other memory operations, resulting in persistent changes to the cache state. Using these observations, we demonstrate (among other attack variants) that secret information accessed by mis-speculated instructions can change the order of bound-to-retire loads. Load timing changes can therefore leave secret-dependent changes in the cache, even in the presence of invisible speculation mechanisms. We show that this problem is not easy to fix: Speculative interference converts timing changes to persistent cache-state changes, and timing is typically ignored by many cache-based defenses. We develop a framework to understand the attack and demonstrate concrete proof-of-concept attacks against invisible speculation mechanisms. We provide security definitions sufficient to block speculative interference attacks; describe a simple defense mechanism with a high performance cost; and discuss how future research can improve its performance., Comment: Updated CR Version

Adding new hardware features to a cloud computing server requires testing both the functionalities and the performance of the new hardware mechanisms. However, commonly used cloud computing server workloads are not well-represented by the SPEC integer and floating-point benchmark and Parsec suites typically used by the computer architecture community. Existing cloud benchmark suites for scale-out or scale-up computing are not representative of the most common cloud usage, and are very difficult to run on a cycle-accurate simulator that can accurately model new hardware, like gem5. In this paper, we present PALMScloud, a suite of cloud computing benchmarks for performance evaluation of cloud servers, that is ready to run on the gem5 cycle-accurate simulator. We demonstrate how our cloud computing benchmarks are used in evaluating the cache performance of a new secure cache called Newcache as a case study. We hope that these cloud benchmarks, ready to run on a dual-machine gem5 simulator or on real machines, can be useful to other researchers interested in improving hardware micro-architecture and cloud server performance.

This paper investigates into the effect of dividend increase announcements to US and China financial market by selecting 30 stocks in Shanghai Stock Exchange and 40 stocks in New York Stock Exchange during the period of 2006 to 2013. The purpose of this research is analyzing the effects of events in short-term event window, which are 5-day event window, 11-day event window, 21-day event window and 61-day event window. We find out that most firms in both China and US financial markets are positively related to the dividend increase announcement but US financial companies seems to have slower reaction speed and longer reaction period. In conclusion, dividend increase announcements have positive abnormal effects to both US and China financial markets but the two markets react increase differently. Investors can only obtain abnormal returns in 30 days after the dividend announcements in China financial markets and can only obtain abnormal returns 10 days after the dividend announcements in US financial markets., Includes abstract and appendices., Includes bibliographical references (p. 32-33).

This article is a quantitative research analysis that aims to compare the career orientations of Swedish and Chinese undergraduates involved in an International Social Work Program as an effort to provide insight for social work educators. Use of the Social Cognitive Career Theory (SCCT) was applied to analyze the data. A mail questionnaire was issued to 51 students studying at the University of Gävle (Sweden). The results show that fifty percent of Swedish students and fifty percent Chinese students would prefer to work in the government and/or public sector. Also, the study participants agree that they will be influenced by career development opportunities in future work, and that the most substantial difference between the nationalities is that Chinese students tend to be more influenced by the social expectations of immediate parents, relatives, and friends, than their Swedish counterparts. Finally, this article lists a few recommendations geared towards helping social work educators provide more relevant links between the school’s curriculum and anticipated job requirements, and bridging more opportunities for student internship positions.

We propose an ultracompact parallel label-free biosensor based on concentric silicon-on-insulator microring resonators. Our numerical studies show that the proposed biosensor offers higher sensitivity and a larger detection area than conventional single-ring-based sensors. We analyze the detection sensitivity of the DNA hybridization by immobilizing the probe on top of the double concentric ring resonators and in the ring-ring air gap. For a concentric double-ring system with an area of 27: 646 mu m(2), the bulk detection sensitivity is 683 nm RIU-1. The sensitive uniformity of the system related to the optical field distribution is also studied., QC 20110310 Asia Optical Fiber and Optoelectronic Exposition and Conference 2008, Shanghai, PEOPLES R CHINA, OCT 30-NOV 02, 2008

In this paper, we experimentally demonstrate continuously tunable pulse propagation in SOI microring resonators with mutual mode coupling, which is induced by nanosized gratings along the ring sidewalls. In the presence of the mutual mode coupling, dispersion-induced group delay is investigated in single-waveguide and double-waveguide coupled single-ring resonators, respectively. In particular, pulse propagation in the drop channel of a double-waveguide coupled resonator is observed and exhibits continuously tunable delay and advancement features in the experiment. Delayed pulses are also obtained at the reflection port of a single-waveguide coupled resonator due to the existence of backward whispering gallery mode (WGM)., QC 20100525

We build a model to study the sidewall corrugation of fabricated silicon microrings and investigate its impact on the spectral response of the resonator system. From the scanning electron microscope images, the sidewall corrugation can be engineered into certain periodicity and characterized into a group of rectangular gratings, or quasi-gratings, which in turn generate mutual mode coupling between the forward- and backward-propagating modes in the ring and lead to resonance splitting. We find that the reflectivity of the quasi-gratings is proportional to the mutual coupling strength and the resonance splitting only occurs at the resonances where high reflectivity takes place. The model agrees well with the experimental measurements and provides some guideline in applying mutual mode coupling for various functions in the field of optics., QC 20100525

We experimentally demonstrate resonance-splitting and enhanced notch as a result of the mutual mode coupling in silicon ring resonator. Dense wavelength conversions, wavelength multicasting and optical up-converter are demonstrated in the split silicon ring resonator. A temporal differentiator and format conversion from NRZ to AMI are demonstrated utilizing the filtering effect of ring resonator. Photonic phase shifter and delay line are demonstrated utilizing the phase shift of ring resonator., QC 20110318

We propose and experimentally demonstrate a temporal differentiator in optical field based on a silicon microring resonator with a radius of 40 mu m. The microring resonator operates near the critical coupling region, and can take the first order derivative of the optical field. It features compact size thus is suitable for integration with silicon-on-insulator (SOI) based optical and electronic devices. The performance of this optical differentiator is tested using signals with typical shapes such as Gaussian, sinusoidal and square-like pulses at data rates of 10 Gb/s and 5 Gb/s., QC 20100525

We experimentally demonstrate optically tunable delay fine in a silicon microring resonator with a 20-mu m radius. The delay-tuning mechanism is based on the red shift of the resonance induced by thermal nonlinear effect. We investigate the delay performance of three modulation formats-non-return-to-zero (NRZ), return-to-zero (RZ), and differential phase-shift keying (DPSK) signals at different data rates. Tunable delay is achieved by controlling the power of the continuous-wave (CW) pump with very low tuning threshold, which could be used in microring-resonator-based slow-light structure., QC 20100923. Uppdaterad från in press till published (20100923).

We theoretically study the group-delay characteristics of a silicon microring resonator based on the coupled mode theory, and experimentally demonstrate error-free operations of an on-chip delay line using a silicon-on-insulator (SOI) microring resonator with a 20-mu m radius. Four signals of different modulation formats are examined at 5 Gb/s, including return-to-zero (RZ), carrier-suppressed return-to-zero (CSRZ), return-to-zero duobinary (RZ-DB), and return-to-zero alternate-mark-inversion (RZ-AMI). Bit error rate (BER) measurements show that the maximal delay times with error-free operations are 80,95, 110, and 65 ps, respectively, corresponding to a fractional group delay of similar to 0.4, similar to 0.5, similar to 0.55, and similar to 0.35. The differences in delay and signal degradations have been investigated based on the signal spectra and pattern dependences. Although the delays are demonstrated in a single ring resonator, the analysis is applicable in slow-light resonance structures such as all-pass filters (APF) and coupled resonator optical waveguides (CROW)., QC 20100525

We experimentally demonstrate all-optical wavelength conversions in a 10 mu m radius resonance-split silicon microring resonator based on free carrier dispersion effect. The split resonance is caused by the mutual coupling between the two countertraveling modes inside the ring resonator. Dense wavelength conversions are performed at data rates from 500 Mbytes/s to 5 Gbytes/s and a dual-channel wavelength multicasting is realized at a data rate of 1.25 Gbytes/s. The resonance splitting phenomenon opens up opportunities to convert more closely spaced wavelengths, thus effectively increasing the system capacity., QC 20100525

We experimentally demonstrate the wavelength conversion based on free carrier dispersion effect induced by non-linear absorption in a silicon micro-ring resonator. The pump and signal wavelengths are set by the split resonances spaced by 0.4 nm induced by deformation in the coupling region. The results of 500 Mbps and 1Gbps are presented., QC 20141021

We demonstrate a microwave-photonic phase shifter based on a 20-ÎŒm-radius silicon microring resonator, providing a tunable phase shift for a 20-GHz signal in a range of 0-4.6 rad., QC 20110315

We report various all-optical signal processing functions using silicon microring resonators with a 450×250-nm cross section. These demonstrations include slow-light delay of phase-modulated data and microwave photonic signal, wavelength conversion/multicasting, format conversions, and optical differentiation., QC 20141017

We propose and numerically verify a novel scheme of all-optical format conversion from non-return-to-zero (NRZ) to binary phase-shift-keying (BPSK) at 160 Gb/s using cascaded microring resonators (CMRR) on a single silicon chip. The conversion is based on large phase shift and flattened intensity-response characteristics in the CMRR. A continuous-wave light experiences different phase shifts controlled by the power of an input NRZ signal with an similar to 8.8-dB extinction ratio, while maintaining approximately the constant intensity. All-optical format conversion from NRZ to quadrature phase-shift-keying (QPSK) is also demonstrated based on parallel NRZ/BPSK converters in a Mach-Zehnder interferometer structure., QC 20100525