Search

BackgroundThe impact of hospital volume on the long-term survival of esophageal squamous cell carcinoma (ESCC) has not been well assessed in China, especially for stage I–III stage ESCC. We performed a large sample size study to assess the relationships between hospital volume and the effectiveness of ESCC treatment and the hospital volume value at the lowest risk of all-cause mortality after esophagectomy in China.AimTo investigate the prognostic value of hospital volume for assessing postoperative long-term survival of ESCC patients in China.MethodsThe date of 158,618 patients with ESCC were collected from a database (1973–2020) established by the State Key Laboratory for Esophageal Cancer Prevention and Treatment, the database includes 500,000 patients with detailed clinical information of pathological diagnosis and staging, treatment approaches and survival follow-up for esophageal and gastric cardia cancers. Intergroup comparisons of patient and treatment characteristics were conducted with the X2 test and analysis of variance. The Kaplan-Meier method with the log-rank test was used to draw the survival curves for the variables tested. A Multivariate Cox proportional hazards regression model was used to analyze the independent prognostic factors for overall survival. The relationship between hospital volume and all-cause mortality was assessed using restricted cubic splines from Cox proportional hazards models. The primary outcome was all-cause mortality.ResultsIn both 1973-1996 and 1997-2020, patients with stage I-III stage ESCC who underwent surgery in high volume hospitals had better survival than those who underwent surgery in low volume hospitals (both PConclusionHospital volume can be used as an indicator to predict the postoperative survival of ESCC patients. Our results suggest that the centralized management of esophageal cancer surgery is meaningful to improve the survival of ESCC patients in China, but the hospital volume should preferably not be higher than 1027 cases/year.Core tipHospital volume is considered to be a prognostic factor for many complex diseases. However, the impact of hospital volume on long-term survival after esophagectomy has not been well evaluated in China. Based on a large sample size of 158,618 ESCC patients in China spanning 47 years (1973-2020), We found that hospital volume can be used as a predictor of postoperative survival in patients with ESCC, and identified hospital volume thresholds with the lowest risk of death from all causes. This may provide an important basis for patients to choose hospitals and have a significant impact on the centralized management of hospital surgery.

BackgroundThis study was intended to construct a brand new prognostic nomogram after combine clinical and pathological characteristics to increases prognostic value in patients with esophageal squamous cell carcinoma.MethodsA total of 1,634 patients were included. Subsequently, the tumor tissues of all patients were prepared into tissue microarrays. AIPATHWELL software was employed to explore tissue microarrays and calculate the tumor-stroma ratio. X-tile was adopted to find the optimal cut-off value. Univariate and multivariate Cox analyses were used to screen out remarkable characteristics for constructing the nomogram in the total populations. A novel prognostic nomogram with clinical and pathological characteristics was constructed on the basis of the training cohort (n=1,144). What’s more performance was validated in the validation cohort (n=490). Clinical-pathological nomogram were assessed by concordance index, time-dependent receiver operating characteristic, calibration curve and decision curve analysis.ResultsThe patients can divide into two groups with cut-off value of 69.78 for the tumor-stroma ratio. It is noteworthy that the survival difference was noticeable (PPConclusionsAs evidently revealed by the research findings, tumor-stroma ratio is an independent prognostic factor in patients with esophageal squamous cell carcinoma. The clinical-pathological nomogram has an incremental value compared TNM stage in predicting overall survival.

Purpose Some studies indicated that gender is associated with prognostic of cancer, However, currently the prognostic value of gender for gastric cardia adenocarcinoma (GCA) survival is unclear. The aim of our study is to reveal the influence of gender on the prognosis of patients with GCA. Patients and methods A total of 42,345 cases Chinese GCA patients were enrolled from our previously established GCA and esophageal cancer databases. The clinicopathological characteristics were retrieved from medical records in hospital. The follow-up was performed through letter, telephone or home interview. Among GCA patients, there were 32,544 (76.9%) male patients with the median age 62 years (range 17–97) and 9,801 (23.1%) female patients with the median age 61 years (range 17–95 years). The Chi-square test and Kaplan–Meier method were used to compare the continuous variables and survival. Cox proportional hazards model was used for competing risk analyses, hazard ratios (HRs) and 95% confidence intervals (CIs) were evaluated. Results Men had shorter GCA-specific survival than women by multivariate analysis (HR 1.114; 95% CI 1.061 to 1.169; P vs. male, P vs. male, P vs. male, P = 0.035). It was worth noting that in patients with stages I, II, III, and IV, female patients survive longer than male patients (P = 0.049; P = 0.011; P P = 0.044, respectively). Conclusion Gender is an independent prognostic factor for patients with GCA. In comparison with men, women have a significantly better outcome. Smoking and drinking may be protective factors for male GCA patients.

Fucosylation is one of the most common modifications of oligo-N-acetyllactosamine (oligo-LacNAc) glycans. However, none of known fucosyltransferases (FucTs) could install the α1,3-linked fucose to the oligo-LacNAc substrates in a site-specific manner. Here, we report a facile and general redox-controlled substrate engineering strategy for the site-specific α1,3-fucosylation of complex glycans containing multiple LacNAc units. This strategy takes advantage of an operationally simple oxidation enzyme module by using galactose oxidase (GOase) to convert the LacNAc unit into oxidized C6'-aldehyde LacNAc sequence, which is not a good substrate for recombinant α1,3-FucT from Helicobacter pylori strain 26695 (Hpα1,3FucT), enabling the site-specific α1,3-fucosylation at intact LacNAc sites. The general applicability and robustness of this strategy were demonstrated by the synthesis of a variety of structurally well-defined fucosides of linear and branched O- and N-linked glycans.

In recent years, time-series data have emerged in a variety of application domains, such as wireless sensor networks and surveillance systems. To identify the similarity between time-series data, the Euclidean distance and its variations are common metrics that quantify the differences between time-series data. However, the Euclidean distance is limited by its inability to elastically shift with the time axis, which motivates the development of dynamic time warping (DTW) algorithms. While DTW algorithms have been proven very useful in diversified applications like speech recognition, their efficacy might be seriously affected by the resolution of the time-series data. However, high-resolution time-series data might take up a gigantic amount of main memory and storage space, which will slow down the DTW analysis procedure. This makes the upscaling of DTW analysis more challenging, especially for in-memory data analytics platforms with limited nonvolatile memory space. In this paper, we propose a strategy to downsample time-series data to significantly reduce their size without seriously affecting the precision of the results obtained by DTW algorithms (downsizing without downgrading). In other words, this paper proposes a technique to remove the unimportant details that are largely ignored by DTW algorithms. The efficacy of the proposed technique is verified by a series of experimental studies, where the results are quite encouraging.

Sda antigens are a family of widely distributed histo-blood group antigens sharing the Siaα2,3(GalNAcβ1,4)Gal trisaccharide determinant. Despite their significant biological importance, the synthesis of these complex carbohydrate antigens is challenging. Herein, we report a highly efficient and green synthesis of diverse Sda antigens with different underlying structures and sialic acid forms utilizing 6 multienzyme cascade reaction systems. The multienzyme cascade approach combines sugar nucleotide donor generation enzymes and glycosyltransferases, providing an efficient and convenient approach for the rapid assembly of complex glycans from simple monosaccharide building blocks.

BackgroundThis study compared the survival outcomes of different surgical approaches to determine the optimal approach for gastric cardia adenocarcinoma (GCA) and aimed to standardize the surgical treatment guidelines for GCA.MethodsA total of 7103 patients with GCA were enrolled from our previously established gastric cardia and esophageal carcinoma databases. In our database, when the epicenter of the tumor was at or within 2 cm distally from the esophagogastric junction, the adenocarcinoma was considered to originate from the cardia and was considered a Siewert type 2 cancer. The main criteria for the enrolled patients included treatment with radical surgery, no radio- or chemotherapy before the operation, and detailed clinicopathological information. Follow-up was mainly performed by telephone or through home interviews. According to the medical records, the surgical approaches included transthoracic, thoracoabdominal, and transabdominal approaches. Kaplan–Meier and Cox proportional hazards regression models were applied to correlate the surgical approach with survival in patients with GCA.ResultsThere were marked differences in age and tumor stage among the patients who underwent the three surgical approaches (P P P P ConclusionThoracoabdominal approach and transabdominal approach were shown to be poor prognostic factors. Patients with (locally advanced) GCA may benefit from the transthoracic approach. Further prospective randomized clinical trials are necessary.

Background: Gastric cardia adenocarcinoma (GCA), which has been normalized as type II of adenocarcinoma at esophagogastric junction in western countries. In clinical, most of the GCA patients are lack of early alarming symptoms, more than 90% of GCA patients were diagnosed at advanced stage, resulted in a very poor prognosis, with less than 20% of 5-year survival. Obviously, early detection for GCA plays crucial role in decreasing the high mortality. Metabolomics allows for appraisal of small molecular mass compounds in a biofluid, which may provide a way for finding biomarkers for GCA. Methods: The serum metabolic features of 276 curatively resected GCA patients and 588 healthy control participates were collected from the database of State Key Laboratory of Esophageal Cancer Prevention & Treatment and Henan Key Laboratory for Esophageal Cancer Research of The First Affiliated Hospital of Zhengzhou University to discover the metabolic dysregulation by using the ultraperformance liquid chromatography-mass spectrometry (UPLC-MS). Joint pathway analysis with metabolites identified, survival analysis and auxiliary diagnosis metabolites were discussed in present work. Results: A sum of 200 known differential metabolites were obtained with p＜0.05 and fold change FC≥1.25 or FC≤0.8 by comparison GCA and healthy control participates. 12 metabolites significant correlated with survival (p＜0.05) and 17 metabolites for potential auxiliary diagnosis(FC＞1.5 or FC＜0.67) for GCA. Dysregulated arginine biosynthesis was an important pathway of GCA. 9 differential metabolites of 12-ketolithocholic acid, 2-Hydroxybutanoic acid, Aldosterone, All-trans-13,14-dihydroretinol, Hododeoxycholic acid, L-histidine, Malonic acid, Prostaglandin E2 and Sphingosine were identified as potential metabolic markers for distinguishing the GCA and healthy control (AUC=0.976, sensitivity =0.913, specificity =0.027, optimal cut off value=0.470). Conclusions: This work was first identified 12 metabolites significant correlated with survival and 17 metabolites for potential auxiliary diagnosis for GCA. In addition, arginine biosynthesis pathway metabolism showed important roles in GCA. Results provide the understanding of the molecular difference between GCA and healthy control. The novel plasma biomarkers panel could clearly distinguish GCA patients from the healthy control group. This finding may form the basis for the development of a minimally invasive method for GCA detection.

Du-Zhong tea is a health beverage produced from Eucommia ulmoides leaves (EUL) as raw materials using traditional tea processing and Chinese herbal pieces processing methods. To evaluate the differences between tender leaves and mature leaves of Du-Zhong tea, UPLC-Q-TOF MS was used to analyze the constituents in EUL collected in April and August. A total of 52 compounds, including iridoids, phenylpropanoids, flavonoids, lignans, and other types of compounds were identified. The contents of nine ingredients in the tender and mature leaves of E. ulmoides were determined by HPLC-DAD analysis. The results show that the average contents of the compounds in tender leaves were significantly higher than those in mature leaves. Lastly, the antioxidant and antipancreatic lipase activities of commercial Du-Zhong tea made by leaves of different maturity were evaluated. Du-Zhong tea with d-grade exhibited relatively higher antioxidant, while C-grade exhibited greater lipase inhibitory activities.

It has been considered as a crucial field for the extraction of active ingredients from herbal medicine to use a green and efficient method in the medicinal and food industries. In recent years, deep eutectic solvents (DESs) have been obtaining increase attention in green chemistry area since its sustainability, safety and biodegradability. In this study, an efficient DES composed of choline chloride and L-(+)-ascorbic acid with a molar ratio of 2:1 performed higher efficacy on the extraction of target compounds (including iridoids, phenolic acids and flavonoids) in Eucommia ulmoides leaves than 50% methanol solution. Considering the extraction efficacy and time consumption, microwave-assisted extraction (MAE) was selected and the operational conditions, including power of microwave, liquid/solid ratio and irradiation time were optimized by response surface methodology (RSM). Water was used as anti-solvent to recover ten target analytes from DES with recovery yields of 97.59%, 94.91%, 96.09%, 90.66%, 95.16%, 87.33%, 86.57%, 82.15%, 89.28% and 80.75% for eucommiol (EU), aucubin (AU), geniposidic acid (GA), chlorogenic acid (CA), asperuloside (AP), rutin (RU), kaempferol-3-O-sambubioside (KS), isoquercitrin (IQ), kaempferol-3-O-rutinoside (KR) and astragaline (AS), respectively. By combining the DES-based MAE and quantitative analysis of multi-components by single mark (QAMS) methods, the contents of ten compounds in the leaves of Eucommia ulmoides were determined to clarify the relationship between the accumulation of secondary metabolites and the harvest period. It was found that the contents of main ingredients reached the highest during May to October. The period appears to be the best harvest period for Eucommia ulmoides leaves. This study provides a novel strategy for the harvesting, processing, and quality control of the raw materials from Eucommia ulmoides leaves.

Extrachromosomal DNA plays an important role in oncogene amplification in tumour cells and poor outcomes across multiple cancers. However, the function of extrachromosomal DNA in gastric cardia adenocarcinoma (GCA) is very limited. Here, we investigated the availability and function of extrachromosomal DNA in GCA from a Chinese cohort of GCA using whole-genome sequencing (WGS), whole-exome sequencing (WES), and immunohistochemistry. For the first time, we identified the ecDNA amplicons present in most GCA patients, and found that some oncogenes are present as ecDNA amplicons in these patients. We found that oncogene ecDNA amplicons in the GCA cohort were associated with the chromothripsis process and may be induced by accumulated DNA damage due to local dietary habits in the geographic region. Strikingly, we observed diverse correlations between the presence of ecDNA oncogene amplicons and prognosis, where ERBB2 ecDNA amplicons correlated with good prognosis, EGFR ecDNA amplicons correlated with poor prognosis, and CCNE1 ecDNA amplicons did not correlate with prognosis. Large-scale ERBB2 immunohistochemistry results from 1668 GCA patients revealed that there was a positive correlation between the presence of ERBB2 and prognosis in 2-7-year survival; however, there was a negative correlation between the presence of ERBB2 and prognosis in 0-2-year survival. Our observations indicate that the presence of ERBB2 ecDNA in GCA patients may represent a good prognosis marker.

It has been considered as a crucial field for the extraction of active ingredients from herbal medicine to use a green and efficient method in the medicinal and food industries. In this study, a deep eutectic solvent (DES) coupled with microwave-assisted extraction (MAE) was used to extract ten compounds from Eucommia ulmoides leaves. Response surface methodology (RSM) combined with Box-Behnken design (BBD) was used to optimize the extraction conditions using the MAE method. An anti-solvent method efficiently recovered target analytes from the DES. By combining the DES-based MAE and quantitative analysis of multi-components by single mark (QAMS) methods, the contents of ten compounds in the leaves of Eucommia ulmoides were determined to clarify the relationship between the accumulation of secondary metabolites and the harvest period. This study provides a novel strategy for the harvesting, processing, and quality control of the raw materials from Eucommia ulmoides leaves.

We propose a malicious viruses spreading model on complex networks that considers the effects of the resource contribution of healthy neighbors. Due to the difference in resource amount received from the healthy neighbors, the recovery rate of infected nodes is heterogeneous. Through rigorous theoretical analysis and extensive numerical simulations, we find that the virus spreading can be optimally suppressed if each healthy node contributes equal resource to the infected nodes. There is a maximum outbreak threshold and minimum fraction of infected nodes with the optimal strategy of resource contribution. In addition, we find that in a homogeneous network, the strategy of resource contribution can alter the phase transition. If each healthy node contributes relatively evenly, the phase transition is continuous. Whereas, if the recovery resources of infected nodes is mainly relies on nodes with large or small degrees, there is discontinuous phase transition. In heterogeneous network, there is always continuous phase transition.

Smartphones are getting increasingly high-performance with advances in mobile processors and larger main memories to support feature-rich applications. However, the storage subsystem has always been a prohibitive factor that slows down the pace of reaching even higher performance while maintaining good user experience. Despite today’s smartphones are equipped with larger-than-ever main memories, they consume more energy and still run out of memory. But the slow NAND flash based storage vetoes the possibility of swapping—an important technique to extend main memory—and leaves a system that constantly terminates user applications under memory pressure. In this paper, we propose NVM-Swap by revisiting swapping for smartphones with fast, byte-addressable, non-volatile memory (NVM) technologies. Instead of using flash, we build the swap area with NVM, to allow high performance without sacrificing user experience. NVM-Swap supports Lazy Swap-in , which can reduce memory copy operations by giving the swapped out pages a second chance to stay in byte-addressable NVM backed swap area. To avoid fast worn-out of certain NVM, we also propose Heap-Wear , a wear leveling algorithm that distributes writes in NVM more evenly. Evaluation results based on the Google Nexus 5 smartphone show that our solution can effectively enhance smartphone performance and achieve better wear-leveling of NVM.

Mobile applications are becoming increasingly feature-rich and powerful, but also dependent on large main memories, which consume a large portion of system energy, especially for devices equipped with 4/6 GB DRAM. Swapping inactive DRAM pages to byte-addressable, non-volatile memory (NVRAM) is a promising solution to this problem. However, most NVRAMs have limited write endurance and the current victim pages selecting algorithm does not aware it. Therefore, to make it practical, the design of an NVRAM based swapping system must also consider endurance. In this paper, we target at prolonging the lifetime of NVRAM based swap area in mobile devices by reducing the write activities to NVRAM based swap area. Different from traditional wisdom, such as wear leveling and hot/cold data identification, we propose to build a system called n Code, which exploits the fact that code pages are easy to identify, read-only, and therefore a perfect candidate for swapping. Utilizing NVRAM’s byte-addressability, we support execute-in-place (XIP) of the code pages in the swap area, without copying them back to DRAM based main memory. Experimental results based on the Google Nexus 5 smartphone show that n Code can effectively prolong the lifetime of NVRAM under various workloads.

Mobile systems such as smartphones and tablets are re-adopting swapping–a mature but rarely used OS feature–to extend memory capacity without adding more DRAM, especially low-end devices. This resurgence of swapping in mobile systems has inspired both traditional “off-the-rack” schemes and new approaches based on compression and new hardware. Their vastly different designs, however, make them difficult for system designers to measure, compare and revise. In this paper, we first propose an evaluation framework, SwapBench, to appraise swap schemes and focus on two important but overlooked metrics: application launch and switch. And cross-validation with microbenchmarks shows that SwapBench is accurate. Then, we present the first comprehensive evaluation from three dimensions: system architecture, application launch time and application switch delays, to understand and summarize the impacts of swapping in mobile systems. Finally, based on the findings from SwapBench, we give our conclusion and suggestions of different approaches to swapping in mobile systems.

Ketamine is an N-methyl-d-aspartate receptor antagonist that is usually used clinically as a racemic mixture. Its two enantiomers exhibit different pharmacological activities. To determine whether the enantiomers have different pharmacokinetic profiles, a chiral liquid chromatography-tandem mass spectrometry method was developed and validated for the determination of ketamine enantiomers in dog plasma. The enantiomers of ketamine were extracted from 50 μL of plasma by methyl tert-butyl ether. Adequate chromatographic retention and baseline resolution of the enantiomers were achieved within a runtime of 5 min on a chiral column coated with polysaccharide derivatives, using a gradient mobile phase of acetonitrile and 10 mm ammonium bicarbonate aqueous solution. Ketamine enantiomers were detected by mass spectrometry with multiple reaction monitoring mode using the transitions of m/z 238.3 → 125.9 for the analytes and m/z 237.1 → 194.1 for carbamazepine (internal standard). The method was linear over the concentration range from 0.5 to 500 ng/mL for each enantiomer. The lower limit of quantification (LLOQ) for each enantiomer was 0.5 ng/mL. The intra- and inter-day precision was

Phase change memory (PCM) is considered as a promising alternative of DRAM-based main memory in embedded systems. A PCM cell can be dynamically programmed to be in either multiple-level cell(MLC) mode or single-level cell(SLC) mode. With this morphable feature, we can utilize the high-density of MLC and low-latency of SLC, to satisfy various memory requirements of specific applications in embedded systems. However, compared to its SLC counterpart, the lifetime of MLC is limited.To address this issue, this paper proposes a simple and effective wear-leveling technique, named Mixer, to enhance the lifetime of morphable PCM considering the program specific features. We first build an Integer Linear Programming(ILP) formulation to dynamically configure the optimal SLC/MLC partition in morphable PCM, and produce the best data allocation for each variable to achieve a balanced write distribution in morphable PCM with low memory access cost. The basic idea is to allocate low-latency SLC and high-density MLC cells for write intensive variables and other ordinary variables, respectively. We then propose a polynomial time algorithm to achieve near-optimal results. The evaluation results show that the proposed technique can effectively improve the lifetime of morphable PCM in embedded systems compared with previous work.

Smartphones are becoming increasingly energy-hungry to support feature-rich applications, posing a lot of pressure on battery lifetime and making energy consumption a non-negligible issue. In particular, dynamic random access memory (DRAM)-based main memory subsystem is a major contributor to the energy consumption of mobile devices. In this paper, we propose direct read (DR). Swap, an energy-efficient in-memory paging design to reduce energy consumption in smartphones. In DR. Swap, we adopt emerging energy-efficient nonvolatile memory (NVM) and use it as the swap area. Utilizing NVMs byte-addressability, we propose DR which guarantees zero memory copy for read-only requests when accessing a page in swap area. To better understand the energy consumption of swapping, we build an energy model to analyze the energy consumption of different paging architectures. We evaluate DR. Swap based on the Google Nexus 5 smartphone, experimental results show that our technique can reduce more than 50% energy consumption compared to DRAM backed swapping.

Virtualization offers significant benefits, such as better isolation and security for mobile systems. However, the limited amount of memory and virtualization’s memory-demanding nature make it challenging to virtualize mobile systems efficiently. In this paper, we utilize morphable resistive memories to design a high-performance mobile system with an extensible memory space. With morphable resistive memories, a simple and effective page management technique, Balloonfish, is proposed to convert the memory cell state between multilevel and single-level for achieving a balance between performance and memory space. First, an application-specific page allocation is proposed for managing morphable resistive memories in virtualized mobile systems. Besides, we use a balloon-style algorithm to balance memory allocation among multiple virtual machines. Our evaluation based on the Samsung Exynos 5250 system-on-chip with various real Android applications shows that our system achieves 28.63% performance improvement compared with the baseline scheme.

Recently, augmented reality technologies are debuting to the mainstream markets. Simultaneous Localization and Mapping (SLAM), which serves as the core to drive augmented reality, enables mobile devices to understand "the reality" by recognizing and understanding the surrounding space. However, enabling SLAM on mobile devices still faces challenges due to computation and power limitations. Thus, this paper explores the microarchitecture design space of visual SLAM. First, we conduct a characterization to make the image signal processing stage in the front-end image acquisition stage configurable and explore SLAM's sensitivity to individual stages. Our characterization shows that only demosaicing, gamma compression, and denoising in the image signal processing process have significant influences on SLAM's accuracy. Thus, the image signal processor in SoC in traditional mobile devices can be replaced by an approximation logic to save hardware overhead and energy. Second, we propose a new CGRA architecture, called SL-CGRA, specially tailored to SLAM's workload characteristics. It features a two-level memory design, which includes data-redirection layer support for on-chip memory, and an efficient CGRA memory controller stacked through Through-Via Silicon (TSV) for highly efficient off-chip memory access. Besides, PEs in SL-CGRA support different execution modes to exploit the data-level parallelism and task-level parallelism of SLAM. Our evaluation results show that SL-CGRA achieves good performance and energy efficiency.

We have witnessed the tremendous success of deep neural networks. However, this success comes with the considerable computation and storage costs which make it difficult to deploy these networks directly on resource-constrained embedded systems. To address this problem, we propose TaiJiNet, a binary-network-based framework that combines binary convolutions and pointwise convolutions, to reduce the computation and storage overhead while maintaining a comparable accuracy. Furthermore, in order to provide TaiJiNet with more flexibility, we introduce a strategy called partial binarized convolution to efficiently balance network performance and accuracy. We evaluate TaiJiNet with the CIFAR-10 and ImageNet datasets. The experimental results show that with the proposed TaiJiNet framework, the binary version of AlexNet can achieve 26x compression rate with a negligible 0.8% accuracy drop when compared with the full-precision AlexNet.

Recent years have seen an exploration of data volumes from a myriad of IoT devices, such as various sensors and ubiquitous cameras. The deluge of IoT data creates enormous opportunities for us to explore the physical world, especially with the help of deep learning techniques. Traditionally, the Cloud is the option for deploying deep learning based applications. However, the challenges of Cloud-centric IoT systems are increasing due to significant data movement overhead, escalating energy needs, and privacy issues. Rather than constantly moving a tremendous amount of raw data to the Cloud, it would be beneficial to leverage the emerging powerful IoT devices to perform the inference task. Nevertheless, the statically trained model could not efficiently handle the dynamic data in the real in-situ environments, which leads to low accuracy. Moreover, the big raw IoT data challenges the traditional supervised training method in the Cloud. To tackle the above challenges, we propose In-situ AI, the first Autonomous and Incremental computing framework and architecture for deep learning based IoT applications. We equip deep learning based IoT system with autonomous IoT data diagnosis (minimize data movement), and incremental and unsupervised training method (tackle the big raw IoT data generated in ever-changing in-situ environments). To provide efficient architectural support for this new computing paradigm, we first characterize the two In-situ AI tasks (i.e. inference and diagnosis tasks) on two popular IoT devices (i.e. mobile GPU and FPGA) and explore the design space and tradeoffs. Based on the characterization results, we propose two working modes for the In-situ AI tasks, including Single-running and Co-running modes. Moreover, we craft analytical models for these two modes to guide the best configuration selection. We also develop a novel two-level weight shared In-situ AI architecture to efficiently deploy In-situ tasks to IoT node. Compared with traditional IoT systems, our In-situ AI can reduce data movement by 28-71%, which further yields 1.4X-3.3X speedup on model update and contributes to 30-70% energy saving.

Aim: To investigate the mechanisms underlying the isomer-selective distribution of 3- n -butylphthalide (NBP) hydroxylated metabolites, 3-hydroxy-NBP (3-OH-NBP) and 10-hydroxy-NBP (10-OH-NBP), across the blood brain barrier (BBB). Methods: After oral administration of NBP (20 mg/kg) to rats, the pharmacokinetics of two major hydroxylated metabolites, 3-OH-NBP and 10-OH-NBP, in plasma and brains were investigated. Plasma and brain protein binding of 3-OH-NBP and 10-OH-NBP was also assessed. To evaluate the influences of major efflux transporters, rats were pretreated with the P-gp inhibitor tariquidar (10 mg/kg, iv) and BCRP inhibitor pantoprazole (40 mg/kg, iv), then received 3-OH-NBP (12 mg/kg, iv) or 10-OH-NBP (3 mg/kg, iv). The metabolic profile of NBP was investigated in rat brain homogenate. Results: After NBP administration, the plasma exposure of 3-OH-NBP was 4.64 times that of 10-OH-NBP, whereas the brain exposure of 3-OH-NBP was only 11.8% of 10-OH-NBP. In the rat plasma, 60%±5.2% of 10-OH-NBP was unbound to proteins versus only 22%±2.3% of 3-OH-NBP being unbound, whereas in the rat brain, free fractions of 3-OH-NBP and 10-OH-NBP were 100%±9.7% and 49.9%±14.1%, respectively. In the rats pretreated with tariquidar and pantoprazole, the unbound partition coefficient K p ,uu of 3-OH-NBP was significantly increased, while that of 10-OH-NBP showed a slight but not statistically significant increase. Incubation of rat brain homogenate with NBP yielded 3-OH-NBP but not 10-OH-NBP. Conclusion: The isomer-selective distribution of 10-OH-NBP and 3-OH-NBP across the BBB of rats is mainly attributed to the differences in plasma and brain protein binding and the efflux transport of 3-OH-NBP. The abundant 10-OH-NBP is not generated in rat brains.

Spirometra erinaceieuropaei is an important diphyllobothriid cestode whose larval plerocercoid (sparganum) can parasitise humans, causing sparganosis. In China, sparganosis is an important foodborne parasitic zoonosis and poses a serious threat to human health. However, our knowledge of the taxonomic classification of S. erinaceieuropaei and its relationship with other diphyllobothriid cestodes, such as Diphyllobothrium, remains fragmentary. In this study, we determined the complete sequences of cytochrome b (cytb) and cytochrome c oxidase subunit I (cox1) genes from 61 isolates of S. erinaceieuropaei collected from 11 geographical locations in eastern China and compared them to the sequences of Diphyllobothrium species from the GenBank database. Then, we evaluated the genetic variation of S. erinaceieuropaei and the phylogenetic relationship between S. erinaceieuropaei and Diphyllobothrium. Both negative Tajima's D and Fu's FS values of neutrality tests and unimodal curve of mismatch distribution analyses supported demographic population expansion. Bayesian skyline plot (BSP) analyses indicated that a population expansion occurred in the Pleistocene. Phylogenetic analysis based on cytb, cox1, and cytb+cox1 under all three methods tested (neighbour-joining, maximum parsimony, and maximum likelihood) suggested that the S. erinaceieuropaei isolates from eastern China and the genus Diphyllobothrium are separate monophyletic groups. Furthermore, the uncorrelated log-normal relaxed molecular-clock model suggested that divergence between Spirometra and Diphyllobothrium began in the late Miocene.

Electrolytic Capacitor (E-cap) is one of the lifetime bottlenecks in power electronic converters. In the last two decades, various active power decoupling circuits have been proposed to improve the reliability of the DC link by eliminating the DC-link E-caps. However, additional active devices could change the stresses of the existing converters, whether the system reliability is improving or not is still an open question. This paper investigates the reliability of the single-phase H-bridge inverter with active power decoupling circuit. The parameter variations in IGBT lifetime model are considered. The Weibull distribution of the key components is obtained from Monte Carlo analysis, and the reliability of the whole system is estimated by the system Reliability Block Diagram (RBD) method. As a case study, 2 kW single-phase H-bridge inverter with passive E-caps and active power decoupling circuits are presented. It is shown that the active power decoupling method is applied to H bridge inverter, and the lifetime of decoupling capacitor can be improved significantly, but it has different effects on system reliability in different applications. In addition, the difference on system reliability of fixed parameter and parameter variations is shown in conclusions.

With high-performance mobile processors and large main memory, smartphones are now integrated with more applications and richer functionality than ever. This poses larger memory and storage space demands, however, most mobile systems have limited memory space, which in turn affects user satisfaction. For example, application response time could become longer due to limited memory capacity. Swapping is an effective way to extend memory capacity, but often lead to poor performance in smartphones. In this paper, we propose predictive process-level swapping that predicts the most rarely used (MRU) applications and dynamically swaps processes ahead-of-time. Trace-based evaluations show up to 30% increase in application launch performance compared to the worst case brought by flash-based swap. The system could retains more than 50% processes in the background with energy cost of no more than 0.03% of battery life for each trigger.

Morinidazole is a novel 5-nitroimidazole antimicrobial drug that undergoes extensive metabolism in humans via N + -glucuronidation ( N + -glucuronide of S -morinidazole [M8-1] and N + -glucuronide of R -morinidazole [M8-2]) and sulfation (sulfate conjugate of morinidazole [M7]). Our objectives were to assess the effects of renal impairment on the pharmacokinetics (PK) of morinidazole and to elucidate the potential mechanisms. In this parallel-group study, healthy subjects and patients with severe renal impairment received an intravenous infusion of 500 mg of morinidazole. Plasma and urine samples were collected and analyzed. The areas under the plasma concentration-time curves (AUC) for M7, M8-1, and M8-2 were 15.1, 20.4, and 17.4 times higher, respectively, in patients with severe renal impairment than in healthy subjects, while the AUC for morinidazole was 1.5 times higher. The urinary recovery of the major metabolites was not significantly different between the two groups over 0 to 48 h, but the renal clearances of M7, M8-1, and M8-2 in patients were 85.3%, 92.5%, and 92.2% lower, respectively. In vitro transporter studies revealed that M7 is a substrate for organic anion transporter 1 (OAT1) and OAT3 ( K m = 28.6 and 54.0 μM, respectively). Only OAT3 transported M8-1 and M8-2. Morinidazole was not a substrate for the transporter-transfected cells examined. These results revealed that the function or activity of renal uptake transporters might be impaired in patients with severe renal impairment, which accounted for dramatically increased plasma exposure and reduced renal clearance of the conjugated metabolites of morinidazole, the substrates of renal transporters in patients. It will help clinicians to adjust the dose in patients with severe renal impairment and to predict possible transporter-based drug-drug interactions.

Morinidazole is a 5-nitroimidazole drug. Its sulfate conjugate M7 was a sensitive substrate of organic anion transporter 1 (OAT1) and OAT3, whereas N+-glucuronides M8-1 and M8-2 were only OAT3 substrates. In chronic renal failure (CRF) patients, plasma exposures of the three conjugates increased by 15-fold, which were also found in 5/6 nephrectomized (5/6 Nx) rats in this study. Although the transcriptions of Oat1 and Oat3 in 5/6 Nx rat kidneys decreased by 50%, no difference was observed on the three conjugate uptakes between control and 5/6 Nx rat kidney slices. Thus, the highly elevated endogenous uremic toxins in 5/6 Nx rats and humans, namely, 3-carboxy-4-methyl-5-propyl-2-furanpropionate (CMPF), hippuric acid (HA), and indoxyl sulfate (IS), were considered as influential factors. In rat kidney slices, the uptake of M7, M8-1, and M8-2 was dose dependently reduced by HA and IS, whose plasma concentrations were elevated 5 times in 5/6 Nx rats. In OAT3-overexpressed cells, the three conjugate uptakes were inhibited by CMPF, HA, and IS with IC50 values of 19.2, 87.4, and 222 μM (M7); 8.53, 39.4, and 161 μM (M8-1); and 6.75, 24.1, and 78.3 μM (M8-2), respectively. In OAT1-overexpressed cells, CMPF, HA, and IS showed weak inhibition on M7 uptake with IC50 values of 187, 162, and 200 μM, correspondingly. Results suggest that the reduced mRNA expression of renal transporters in CRF patients may not influence the activities of these transporters. However, accumulated uremic toxins may inhibit the transporters, particularly OAT3, leading to plasma exposure changes of relevant substrates.

GSK-3β的过度表达可导致人脑神经细胞内Tau蛋白的过磷酸化, 从而介导阿尔兹海默病(Alzheimer's disease, AD)的发生. 本文旨在研究GSK-3β的马来酰胺类抑制剂的三维定量构效关系(3D-QSAR)及新抑制剂分子与GSK-3β的作用机制. 采用基于R基团搜索技术的Topomer CoMFA建立了49个马来酰胺类GSK-3β抑制剂的3D-QSAR模型, 并用包括25个样本的测试集验证模型的外部预测能力. 所得优化模型的拟合、交互验证以及外部验证的复相关系数分别为0.928, 0.790和0.725. 采用Topomer search在ZINC分子数据库中进行虚拟搜索, 设计了28个可能具有更高活性的新抑制剂. 借助Surflex-dock分子对接研究了新抑制剂与GSK-3β作用模式与机制. 结果显示, 新抑制剂与GSK-3β的Asp133, Tyr134, Val135和Pro136等位点作用显著.

Embedded sensor nodes are sensitize to battery lifetime and evidences show that DRAM-based main memory subsystem is the major contributor of the energy consumption of embedded sensor nodes. Due to the high density, byte-addressability, and low standby power consumption, non-volatile random access memories (NVRAMs), such as PRAM and STT-RAM, become promising main memory alternatives in embedded sensor nodes. On the other hand, NAND flash memory is widely adopted for storing collected data in embedded sensor nodes. However, both NVRAM and NAND flash memory have limited lifetime, how to optimize the management of NAND flash memory in NVRAM-based embedded sensor nodes while considering the endurance issue becomes quite important. In this chapter, we introduce a write-actively-aware NAND flash memory management scheme to effectively manage NAND flash memory while reducing the write activities to NVRAM-based main memory in embedded sensor nodes. The basic idea is to preserve each bit in flash mapping table, which is stored in NVRAM, from being inverted frequently during the mapping table update process. To achieve this, a two-level mapping mechanism is employed while considering the access behavior of IO requests, and a customized wear-leveling scheme is developed to evenly distribute the writes across the whole mapping table. Evaluation results show that the proposed technique can reduce the write activities significantly and achieve an even distribution of writes in NVRAM with low overhead.

Power line communication is popular at indoor communication and smart home applications, which face a harsh using condition that is of bad impedance match, large signal attenuation, unexpectable noise and channel high time variation. These difficulties set a high threshold for power line communication technology and improve the communication channel simulation to an important place. To predetermine the performance of power line communication productions, a narrow-band communication simulation platform was developed. This platform is based on low voltage communication channel model, and can update the system parameters using on site measured data, which could get a match with the real application environment. Main channel characteristics, such as noise, signal attenuation and frequency respond could be simulated on this platform. Noise generation frequency range of this platform is 0~120MHz, attenuation frequency range is 0~4GHz and attenuation range is 0~120dB. All performance indexes can well cover the measurement boundary of power line communication.

Role of nitrogen-containing organic compounds produced during the plasma process on the subsequent of C 2 H 4 -SCR of NO x over Ag/γ-Al 2 O 3 catalyst at 523 K was studied in a two-stage plasma-facilitated catalysis system. Results showed that nitrogen-containing organic compounds (methyl nitrate, ethyl nitrate, nitro methyl) produced in the plasma reactor can improve the subsequent catalytic performance at lower temperature besides NO 2 and aldehyde; more the amounts of nitrogen-containing organic compounds produced after plasma pretreatment, more the amounts of N 2 produced in the subsequent catalysis process, and N 2 selectivity leveled at about 82%.

Synergy study is of great importance to optimize the two-stage plasma-facilitated catalysis system and also necessary to gain a mechanistic understanding of the overall reaction process. In this paper, the effects of the long-lived chemical products of plasma pretreatment on the subsequent selective catalytic reduction (SCR) of NOx at fixed and changed energy density of dielectric-barrier-discharge plasma reactor were investigated, respectively, for the aim of the synergy study of plasma- facilitated C2H4-SCR (PF-SCR) of NOx over Ag/gamma-Al2O3 catalyst at lower temperature. For comparison, C2H4-SCR of NOx over Ag/gamma-Al2O3 catalyst alone was also considered. Results showed that PF-SCR for NOx removal exhibited obvious synergistic effect at 240degC, and the NOx removal efficiency for plasma alone, C2H4-SCR alone, and PF-SCR were 43.6%, 6.7%, and 90.1%, respectively. In the NO/NO2/C2H4/O2/N2 system, the long-lived chemical products of plasma pretreatment for the synergistic link between plasma and catalysis included NO2, aldehyde (formaldehyde and acetaldehyde), and nitrogen-contained organic compounds (methyl nitrate, nitro methyl, and ethyl nitrate). Among the nitrogen-contained chemical products of plasma pretreatment, the major synergistic link between plasma and catalysis was NO2 at lower input energy density, transferred to the mutual action of NO2 and nitrogen-contained organic compounds over catalyst at higher input energy density.

Emerging non-volatile memory (NVRAM) has been considered as the most promising candidate of DRAM for future main memory design in mobile devices. NVRAMbased main memory exhibits attractive features, such as byteaddressability, low standby power, high density and near DRAM performance. However, the nature of non-volatility makes NVRAM vulnerable to be attacked by malicious programs. Though several data encryption techniques have been proposed to solve this problem, they do not consider the limited resources in mobile devices. To address this issue, in this paper, we propose an energyaware encryption mechanism, named MobiLock, to effectively enhance the security of NVRAM-based main memory in mobile devices. The basic idea is to enhance the encryption and decryption performance by utilizing cache and concurrency mechanisms, respectively. To achieve this, we first develop a cache mechanism to cache the encrypted intermediate data (i.e., PAD) whose plaintexts are updated frequently, for accelerating decryption and reducing reamputation of PAD. We then propose a concurrency mechanism to read the ciphertext in NVRAM and calculate the PAD simultaneously, to reduce the decryption latency. The evaluation results show that our technique can effectively reduce encryption energy consumption and decryption latency, respectively.

Mobile systems such as smartphones and tablets are re-adopting swapping -- a mature but by rarely used OS feature -- to extend memory capacity without adding more DRAM, especially low-end devices. This resurgence of swapping in mobile systems has inspired both traditional "off-the-rack" schemes and new approaches based on compression and new hardware. Their vastly different designs, however, make them difficult for system designers to measure, compare and revise. In this paper, we propose an evaluation framework, SwapBench, to appraise swap schemes and focus on two important but overlooked metrics: application launch and switch. Cross-validation with microbenchmarks shows that SwapBench is accurate. Based on the findings from SwapBench, we further discuss the impacts of different approaches to swapping in mobile systems.

Mobile virtualization provides a secure and reliable mobile ecosystem in many aspects, such as personal privacy, corporate and public security. However, the limited memory resources of mobile systems and virtualization's memory demanding feature post challenges on main memory. Fortunately, non-volatile random access memory(NVRAM) provides opportunities to solve the problem with its attractive features, such as high density, low standby power and so on. However, the long write latency of NVRAM may slow down the performance of write-intensive tasks. Time-critical tasks may end up with unbearable performance. In this paper, we therefore focus on improving the performance of mobile virtualization with NVRAM/DRAM hybrid main memory. To achieve this, we propose NV-CFS, an user-centric virtual machine scheduler built for the ARM/KVM hypervisor. Experimental results based on various pseudo-randomly selected task sets show that NV-CFS can improve the performance more than 30%, and significantly reduce the task deadline miss ratio when compared to the traditional scheduling algorithm.

Similar to traditional NAND flash memory, triple-level cell (TLC) flash memory is used as secondary storage to meet the fast growing demands on storage capacity. TLC flash memory exhibits attractive features such as shock resistance, high density, low cost, non-volatility and low access latency natures. However, TLC flash memory also has some extra limitations, such as write disturbance, low performances and very limited cycles compared to single-level cell (SLC) flash memory. In this paper, we propose a workload-aware flash translation layer, named TLC-FTL, for the TLC/SLC dual-mode enabled flash memory, to improve performance and lifespan of the system. The basic idea is to classify metadata/userdata according to their access pattern, and store the hot/code data in SLC/TLC-mode flash, respectively. TLC-FTL dynamically allocates TLC/SLC capacity based on different workloads. Experimental results show that TLC-FTL can effectively improve the performance and lifetime of the TLC/SLC dual-mode flash memory in embedded systems.

Phase change memory (PCM) is considered as a promising alternative of DRAM-based main memory in embedded systems. A PCM cell can be dynamically programmed to be in either multiple-level cell (MLC) mode or single-level cell (SLC) mode. With this morphable feature, we can utilize the high-density of MLC and low-latency of SLC, to satisfy various memory requirements of specific applications in embedded systems. However, compared to its SLC counterpart, the lifetime of MLC is limited. In particular, a program in embedded systems usually exhibits an extremely unbalanced write pattern, which may accelerate the wear-out of MLC cells in morphable PCM. To address this issue, this paper proposes a simple and effective wear-leveling technique, named Mixer, to enhance the lifetime of morphable PCM considering the program specific features. We first build an Integer Linear Programming (ILP) formulation to produce optimal SLC/MLC partition and data allocation, to achieve a balanced write distribution in morphable PCM with low memory access cost. The basic idea is to allocate fast SLC and MLC cells for write intensive variables and other ordinary variables, respectively. We then propose a polynomial time algorithm to achieve near-optimal results. The evaluation results show that the proposed technique can effectively extend the lifetime of morphable PCM in embedded systems compared with previous work.

Clopidogrel, a thienopyridine antiplatelet prodrug, is metabolized by oxidation to 2-oxo-clopidogrel, followed by conversion to its pharmacologically active thiol metabolite. After oral administration of clopidogrel to humans, two thiol isomers (H3 and H4) are observed in plasma, with similar concentrations, and only H4 is active in humans. In this work, the mechanism of stereoselectivity in the formation and S-methylation of H3 and H4 was investigated in vitro. The two diastereomers of 2-oxo-clopidogrel were epimerized rapidly at physiologic pH. The intrinsic clearance (CLint) for H3 formation from 2-oxo-clopidogrel in human liver microsomes (HLMs) was 3.1-fold higher than that for H4 formation, indicating stereoselective metabolism. Kinetic studies using expressed enzymes demonstrated that the contributions of CYP2B6, CYP2C19, and CYP3A4 to the formation of H4 from 2-oxo-clopidogrel were 18.5%, 26.1%, and 53.5%, respectively. The CLint ratios of H3 formation to H4 formation from 2-oxo-clopidogrel by CYP2B6, CYP2C19, and CYP3A4 were 2.2, 1.0, and 1.7, respectively. In HLMs, H3 and H4 were further S-methylated, and the S-methylation was inhibited by 2,3-dichloromethyl benzylamine, indicating the involvement of thiol S-methyltransferase. The CLint value for the S-methylation of H3 in HLMs was 98.1-fold higher than that for H4. The stereoselective formation of H3 from 2-oxo-clopidogrel and the stereoselective S-methylation of H3 may lead to the similar exposure levels of H3 and H4 previously reported in humans. The epimerization of 2-oxo-clopidogrel and the variations of thiol S-methyltransferase may affect the exposure to H4 in humans.

A rapid, sensitive and convenient LC-MS/MS method was developed for the determination of γ-aminobutyric acid (GABA) in human plasma. d2-γ-Aminobutyric acid (d2-GABA) was synthesized as internal standard (IS). After extraction from human plasma by protein precipitation with acetonitrile, all analytes were separated on a Luna HILIC column (100 mm x 3.0 mm, 3 μm) using an isocratic mobile phase of water: acetonitrile: formic acid (20 : 80 : 0.12) with a flow rate of 0.5 mL x min(-1). Acquisition of mass spectrometric data was performed in multiple reaction monitoring mode (MRM) in positive electrospray ionization using the transitions of m/z 104 --69 for GABA and m/z 106 --71 for d2-GABA. The method was linear in the concentration range of 5.00 to 1 000 ng x mL(-1). The intra- and inter-day precisions were within 9.9%, and accuracy ranged from 99.1% to 104%, within the acceptable limit across all concentrations. The method was successfully applied to a pharmacokinetic study of GABA tablets in healthy Chinese volunteers.

Fasiglifam (TAK-875), a selective G-protein-coupled receptor 40 agonist, was developed for the treatment of type 2 diabetes mellitus; however, its development was terminated in phase III clinical trials because of liver safety concerns. Our preliminary study indicated that intravenous administration of 100 mg/kg of TAK-875 increased the serum total bile acid concentration by 3 to 4 times and total bilirubin levels by 1.5 to 2.6 times in rats. In the present study, we examined the inhibitory effects of TAK-875 on hepatobiliary transporters to explore the mechanisms underlying its hepatotoxicity. TAK-875 decreased the biliary excretion index and the in vitro biliary clearance of d₈-taurocholic acid in sandwich-cultured rat hepatocytes, suggesting that TAK-875 impaired biliary excretion of bile acids, possibly by inhibiting bile salt export pump (Bsep). TAK-875 inhibited the efflux transporter multidrug resistance-associated protein 2 (Mrp2) in rat hepatocytes using 5 (and 6)-carboxy-2',7'-dichlorofluorescein as a substrate. Inhibition of MRP2 was further confirmed by reduced transport of vinblastine in Madin-Darby canine kidney cells overexpressing MRP2 with IC₅₀ values of 2.41 μM. TAK-875 also inhibited the major bile acid uptake transporter Na(+)/taurocholate cotransporting polypeptide (Ntcp), which transports d₈-taurocholic acid into rat hepatocytes, with an IC₅₀ value of 10.9 μM. TAK-875 significantly inhibited atorvastatin uptake in organic anion transporter protein (OATP) 1B1 and OATP1B3 cells with IC₅₀ values of 2.28 and 3.98 μM, respectively. These results indicate that TAK-875 inhibited the efflux transporter MRP2/Mrp2 and uptake transporters Ntcp and OATP/Oatp, which may affect bile acid and bilirubin homeostasis, resulting in hyperbilirubinemia and cholestatic hepatotoxicity.

Virtualization offers significant benefits such as better isolation and security for mobile systems. However, the limited amount of memory and virtualization's memory-demanding nature makes it challenging to virtualize mobile systems efficiently. In this paper, we utilize morphable resistive memories to design a high-performance mobile system with extensible memory space. With morphable resistive memory, we convert the memory cell state between multi-level and single-level to achieve a balance between performance and memory space. Our evaluation based on the Samsung Exynos 5250 SoC with real Android applications shows that our system achieve 27% performance improvement compared with the baseline scheme.

Smartphones are getting increasingly high-performance with advances in mobile processors and larger main memories to support feature-rich applications. However, the storage subsystem has always been a prohibitive factor that slows down the pace of reaching even higher performance while maintaining good user experience. Despite today's smartphones are equipped with larger-than-ever main memories, they consume more energy and still run out of memory. But the slow NAND flash based storage vetoes the possibility of swapping---an important technique to extend main memory---and leaves a system that constantly terminates user applications under memory pressure.In this paper, we revisit swapping for smartphones with fast, byte-addressable, non-volatile memory (NVM) technologies. Instead of using flash, we build the swap area with NVM, to allow high performance without sacrificing user experience. Based on NVM's high performance and byte-addressability, we show that a copy-on-write swap-in scheme can achieve even better performance by avoiding unnecessary memory copy operations. To avoid fast worn-out of certain NVMs, we also propose Heap-Wear, a wear leveling algorithm that more evenly distributes writes in NVM. Evaluation results based on the Google Nexus 5 smartphone show that our solution can effectively enhance smartphone performance and give better wear-leveling of NVM.

Smartphones are becoming increasingly energy-hungry to support feature-rich applications, posing a lot of pressure on battery lifetime and making energy consumption a non-negligible issue. In particular, DRAM is among the most demanding components in energy consumption. In this paper, we propose DR. Swap, an energy-efficient paging design to reduce energy consumption in smartphones. We adopt emerging energy-efficient non-volatile memory (NVM) and use it as the swap area. Utilizing NVM's byte-addressability, we propose direct read which guarantees zero-copy for read-only pages in the swap area. Experimental results based on the Google Nexus 5 smartphone show that our technique can effectively reduce energy consumption.