Engineering Papers⌕ Search

Engineering topics

Li, Dong

Publications and source records attributed to Li, Dong.

At least 55 records · Page 3

Three distinct modes in a surface micro-discharge in atmospheric pressure He + N{sub 2} mixtures

A surface micro-discharge in atmospheric pressure He + N{sub 2} mixtures is studied in this paper with an emphasis on the discharge modes. With the N{sub 2} admixture increasing from 0.1% to 20%, the discharge evolves from a spatially diffuse mode to a filamentary mode during positive half-cycles of the applied voltage. However during the negative half-cycles, an additional patterned mode emerges between the diffuse and the filamentary modes, which has not been reported before to exist in surface micro-discharges. In the diffuse and patterned modes, the plasmas cover almost the entirety of the mesh area during one cycle after plasma ignition in all mesh elements, and the discharge power increases linearly with the applied voltage. In contrast, plasma coverage of the mesh area is only partial in the filamentary mode and the plasma is more unstable with the discharge power increasing exponentially with the applied voltage. As the surface micro-discharge evolves through the three modes, the density of excited species changes significantly, for instance, the density of N{sub 2}{sup +}(B) drops by ∼20-fold from [N{sub 2}] = 0.2% to 20%. The N{sub 2}{sup +}(B) is predicted to be generated mainly through successive processes of Penning ionization by helium metastables and electron-impact excitation of N{sub 2}{sup +}(X), the latter is most responsible for the density decrease of N{sub 2}{sup +}(B) because much more N{sub 2}{sup +}(X) is converted to N{sub 4}{sup +}(X) as the increase of N{sub 2} fraction. Also, the electron density and electron temperature decrease with the discharge mode transition.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Fast Fault Injection and Sensitivity Analysis for Collective Communications, In: 2015 IEEE International Conference on Cluster Computing

The collective communication operations, which are widely used in parallel applications for global communication and synchronization are critical for application's performance and scalability. However, how faulty collective communications impact the application and how errors propagate between the application processes is largely unexplored. One of the critical reasons for this situation is the lack of fast evaluation method to investigate the impacts of faulty collective operations. The traditional random fault injection methods relying on a large amount of fault injection tests to ensure statistical significance require a significant amount of resources and time. These methods result in prohibitive evaluation cost when applied to the collectives. In this paper, we introduce a novel tool named Fast Fault Injection and Sensitivity Analysis Tool (FastFIT) to conduct fast fault injection and characterize the application sensitivity to faulty collectives. The tool achieves fast exploration by reducing the exploration space and predicting the application sensitivity using Machine Learning (ML) techniques. A basis for these techniques are implicit correlations between MPI semantics, application context, critical application features, and application responses to faulty collective communications. The experimental results show that our approach reduces the fault injection points and tests by 97% for representative benchmarks (NAS Parallel Benchmarks (NPB)) and a realistic application (Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS)) on a production supercomputer. Further, we statistically generalize the application sensitivity to faulty collective communications for these workloads, and present correlation between application features and the sensitivity.

Computer Science↗

Ionic switch controls the DNA state in phage λ

We have recently found that DNA packaged in phage λ undergoes a disordering transition triggered by temperature, which results in increased genome mobility. This solid-to-fluid like DNA transition markedly increases the number of infectious λ particles facilitating infection. However, the structural transition strongly depends on temperature and ionic conditions in the surrounding medium. Using titration microcalorimetry combined with solution X-ray scattering, we mapped both energetic and structural changes associated with transition of the encapsidated λ-DNA. Packaged DNA needs to reach a critical stress level in order for transition to occur. We varied the stress on DNA in the capsid by changing the temperature, packaged DNA length and ionic conditions. We found striking evidence that the intracapsid DNA transition is ‘switched on’ at the ionic conditions mimicking those in vivo and also at the physiologic temperature of infection at 37°C. This ion regulated on-off switch of packaged DNA mobility in turn affects viral replication. The results suggest a remarkable adaptation of phage λ to the environment of its host bacteria in the human gut. The metastable DNA state in the capsid provides a new paradigm for the physical evolution of viruses.

59 BASIC BIOLOGICAL SCIENCES↗

Dynamics of charge clouds ejected from laser-induced warm dense gold nanofilms

We report the first systematic study of the ejected charge dynamics surrounding laser-produced 30-nm warm dense gold films using single-shot femtosecond electron shadow imaging and deflectometry. The results reveal a two-step dynamical process of the ejected electrons under the high pump fluence conditions: an initial emission and accumulation of a large amount of electrons near the pumped surface region followed by the formation of hemispherical clouds of electrons on both sides of the film, which are escaping into the vacuum at a nearly isotropic and constant velocity with an unusually high kinetic energy of more than 300 eV. We also developed a model of the escaping charge distribution that not only reproduces the main features of the observed charge expansion dynamics but also allows us to extract the number of ejected electrons remaining in the cloud.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Solid-to-fluid – like DNA transition in viruses facilitates infection

Releasing the packaged viral DNA into the host cell is an essential process to initiate viral infection. In many double-stranded DNA bacterial viruses and herpesviruses, the tightly packaged genome is hexagonally ordered and stressed in the protein shell, called the capsid. DNA condensed in this state inside viral capsids has been shown to be trapped in a glassy state, with restricted molecular motion in vitro. This limited intracapsid DNA mobility is caused by the sliding friction between closely packaged DNA strands, as a result of the repulsive interactions between the negative charges on the DNA helices. It had been unclear how this rigid crystalline structure of the viral genome rapidly ejects from the capsid, reaching rates of 60,000 bp/s. Through a combination of single- molecule and bulk techniques, we determined how the structure and energy of the encapsidated DNA in phage λ regulates the mobility required for its ejection. Our data show that packaged λ -DNA undergoes a solid-to-fluid – like disordering transition as a function of temperature, resultin g locally in less densely packed DNA, reducing DNA – DNA repulsions. This p rocess leads to a sig- nificant increase in genome mobility or fluidity, which facilitates genome release at temperatures close to that of viral infection (37 °C), suggesting a remarkab le physical adaptation of bac- terial viruses to the environment of Escherichia coli cells in a human host.

AFM↗

Solid-to-fluid DNA transition inside HSV-1 capsid close to the temperature of infection

DNA in the human Herpes simplex virus type 1 (HSV-1) capsid is packaged to a tight density. This leads to tens of atmospheres of internal pressure responsible for the delivery of the herpes genome into the cell nucleus. In this study we show that, despite its liquid crystalline state inside the capsid, the DNA is fluid-like, which facilitates its ejection into the cell nucleus during infection. We found that the sliding friction between closely packaged DNA strands, caused by interstrand repulsive interactions, is reduced by the ionic environment of epithelial cells and neurons susceptible to herpes infection. However, variations in the ionic conditions corresponding to neuronal activity can restrict DNA mobility in the capsid, making it more solid-like. This can inhibit intranuclear DNA release and interfere with viral replication. In addition, the temperature of the human host (37 °C) induces a disordering transition of the encapsidated herpes genome, which reduces interstrand interactions and provides genome mobility required for infection.

Sae-Ueng, Udom↗

Reduction of magnetic damping constant of FeCo films by rare-earth Gd doping

Magnetic damping constant (α) is one of the key parameters to determine the critical current density of spin-transfer-torque devices and the switching time of magnetization for ultra-high-frequency devices. In this work, Gd doped FeCo films were fabricated to investigate α based on the ferromagnetic resonance technique. Gd doping not only can efficiently decrease the magnetic inhomogeneity and the extrinsic part of α but also the Landé g-factor and intrinsic part of α. The obtained α was roughly proportional to (g-2){sup 2} and the magnetic anisotropic constant, indicating that the decreased spin-orbit interaction decreases α by Gd doping.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Methane yield phenotypes linked to differential gene expression in the sheep rumen microbiome

Ruminant livestock represent the single largest anthropogenic source of the potent greenhouse gas methane, which is generated by methanogenic archaea residing in ruminant digestive tracts. While differences between individual animals of the same breed in the amount of methane produced have been observed, the basis for this variation remains to be elucidated. To explore the mechanistic basis of this methane production, we measured methane yields from 22 sheep, which revealed that methane yields are a reproducible, quantitative trait. Deep metagenomic and metatranscriptomic sequencing demonstrated a similar abundance of methanogens and methanogenesis pathway genes in high and low methane emitters. However, transcription of methanogenesis pathway genes was substantially increased in sheep with high methane yields. These results identify a discrete set of rumen methanogens whose methanogenesis pathway transcription profiles correlate with methane yields and provide new targets for CH4 mitigation at the levels of microbiota composition and transcriptional regulation.

59 BASIC BIOLOGICAL SCIENCES↗

A Survey Of Architectural Approaches for Managing Embedded DRAM and Non-volatile On-chip Caches

Recent trends of CMOS scaling and increasing number of on-chip cores have led to a large increase in the size of on-chip caches. Since SRAM has low density and consumes large amount of leakage power, its use in designing on-chip caches has become more challenging. To address this issue, researchers are exploring the use of several emerging memory technologies, such as embedded DRAM, spin transfer torque RAM, resistive RAM, phase change RAM and domain wall memory. In this paper, we survey the architectural approaches proposed for designing memory systems and, specifically, caches with these emerging memory technologies. To highlight their similarities and differences, we present a classification of these technologies and architectural approaches based on their key characteristics. We also briefly summarize the challenges in using these technologies for architecting caches. We believe that this survey will help the readers gain insights into the emerging memory device technologies, and their potential use in designing future computing systems.

classification↗

WriteSmoothing: Improving Lifetime of Non-volatile Caches Using Intra-set Wear-leveling

Driven by the trends of increasing core-count and bandwidth-wall problem, the size of last level caches (LLCs) has greatly increased. Since SRAM consumes high leakage power, researchers have explored use of non-volatile memories (NVMs) for designing caches as they provide high density and consume low leakage power. However, since NVMs have low write-endurance and the existing cache management policies are write variation-unaware, effective wear-leveling techniques are required for achieving reasonable cache lifetimes using NVMs. We present WriteSmoothing, a technique for mitigating intra-set write variation in NVM caches. WriteSmoothing logically divides the cache-sets into multiple modules. For each module, WriteSmoothing collectively records number of writes in each way for any of the sets. It then periodically makes most frequently written ways in a module unavailable to shift the write-pressure to other ways in the sets of the module. Extensive simulation results have shown that on average, for single and dual-core system configurations, WriteSmoothing improves cache lifetime by 2.17X and 2.75X, respectively. Also, its implementation overhead is small and it works well for a wide range of algorithm and system parameters.

cache↗

Improving energy efficiency of Embedded DRAM Caches for High-end Computing Systems

With increasing system core-count, the size of last level cache (LLC) has increased and since SRAM consumes high leakage power, power consumption of LLCs is becoming a significant fraction of processor power consumption. To address this, researchers have used embedded DRAM (eDRAM) LLCs which consume low-leakage power. However, eDRAM caches consume a significant amount of energy in the form of refresh energy. In this paper, we propose ESTEEM, an energy saving technique for embedded DRAM caches. ESTEEM uses dynamic cache reconfiguration to turn-off a portion of the cache to save both leakage and refresh energy. It logically divides the cache sets into multiple modules and turns-off possibly different number of ways in each module. Microarchitectural simulations confirm that ESTEEM is effective in improving performance and energy efficiency and provides better results compared to a recently-proposed eDRAM cache energy saving technique, namely Refrint. For single and dual-core simulations, the average saving in memory subsystem (LLC+main memory) on using ESTEEM is 25.8% and 32.6%, respectively and average weighted speedup are 1.09X and 1.22X, respectively. Additional experiments confirm that ESTEEM works well for a wide-range of system parameters.

cache reconfiguration↗

LastingNVCache: A Technique for Improving the Lifetime of Non-volatile Caches

Use of NVM (Non-volatile memory) devices such as ReRAM (resistive RAM) and STT-RAM (spin transfer torque RAM) for designing on-chip caches holds the promise of providing a high-density, low-leakage alternative to SRAM. However, low write endurance of NVMs, along with the write-variation introduced by existing cache management schemes may significantly limit the lifetime of NVM caches. We present LastingNVCache, a technique for improving lifetime of NVM caches by mitigating the intra-set write variation. LastingNVCache works on the key idea that by periodically flushing a frequently-written data-item, the next time the block can be made to load into a cold block in the set. Through this, the future writes to that data-item can be redirected from a hot block to a cold block, which leads to improvement in the cache lifetime. Microarchitectural simulations have shown that LastingNVCache provides 6.36X, 9.79X, and 10.94X improvement in lifetime for single, dual and quad-core systems. Also, its implementation overhead is small and it outperforms a recently proposed technique for improving lifetime of NVM caches.

device lifetime↗