Data Dependence Analysis Research Articles

An integrated task computation and data management scheduling strategy for workflow applications in cloud environments

A workflow is a systematic computation or a data-intensive application that has a regular computation and data access patterns. It is a key to design scalable scheduling algorithms in Cloud environments to address these runtime regularities effectively. While existing researches ignore to join the tasks scheduling and the optimization of data management for workflow, little attention has been paid so far to understand the combination between the two. The proposed scheme indicates that the coordination between task computation and data management can improve the scheduling performance.Our model considers data management to obtain satisfactory makespan on multiple datacenters. At the same time, our adaptive data-dependency analysis can reveal parallelization opportunities. In this paper, we introduce an adaptive data-aware scheduling (ADAS) strategy for workflow applications. It consist of a set-up stage which builds the clusters for the workflow tasks and datasets, and a run-time stage which makes the overlapped execution for the workflows. Through rigorous performance evaluation studies, we demonstrate that our strategy can effectively improve the workflow completion time and utilization of resources in a Cloud environment.

Summary-Based Context-Sensitive Data-Dependence Analysis in Presence of Callbacks

Building a summary for library code is a common approach to speeding up the analysis of client code. In presence of callbacks, some reachability relationships between library nodes cannot be obtained during library-code summarization. Thus, the library code may have to be analyzed again during the analysis of the client code with the library summary. In this paper, we propose to summarize library code with tree-adjoining-language (TAL) reachability. Compared with the summary built with context-free-language (CFL) reachability, the summary built with TAL reachability further contains conditional reachability relationships. The conditional reachability relationships can lead to much lighter analysis of the library code during the client code analysis with the TAL-reachability-based library summary. We also performed an experimental comparison of context-sensitive data-dependence analysis with the TAL-reachability-based library summary and context-sensitive data-dependence analysis with the CFL-reachability-based library summary using 15 benchmark subjects. Our experimental results demonstrate that the former has an 8X speed-up over the latter on average.

Picos: A hardware runtime architecture support for OmpSs

OmpSs is a programming model that provides a simple and powerful way of annotating sequential programs to exploit heterogeneity and task parallelism based on runtime data dependency analysis, dataflow scheduling and out-of-order task execution; it has greatly influenced Version 4.0 of the OpenMP standard. The current implementation of OmpSs achieves those capabilities with a pure-software runtime library: Nanos++. Therefore, although powerful and easy to use, the performance benefits of exploiting fine-grained (pico) task parallelism are limited by the software runtime overheads. To overcome this handicap we propose Picos, an implementation of the Task Superscalar (TSS) architecture that provides hardware support to the OmpSs programming model. Picos is a novel hardware dataflow-based task scheduler that dynamically analyzes inter-task dependencies and identifies task-level parallelism at run-time. In this paper, we describe the Picos Hardware Design and the latencies of the main functionality of its components, based on the synthesis of their VHDL design. We have implemented a full cycle-accurate simulator based on those latencies to perform a design exploration of the characteristics and number of its components in a reasonable amount of time. Finally, we present a comparison of the Picos and Nanos++ runtime performance scalability with a set of real benchmarks. With Picos, a programmer can achieve ideal scalability using aggressive parallel strategies with a large number of fine granularity tasks.

Parallelization of an Encryption Algorithm Based on a Spatiotemporal Chaotic System and a Chaotic Neural Network

Abstract In this paper the results of parallelizing a block cipher based on a spatiotemporal chaotic system and a chaotic neural network are presented. A data dependence analysis of loops was applied in order to parallelize the algorithm. The parallelism of the algorithm is demonstrated in accordance with the OpenMP standard. The efficiency measurements of a parallel algorithm working in standard modes of operation are shown.

Relative quantification of amine-containing metabolites using isobaric N,N-dimethyl leucine (DiLeu) reagents via LC-ESI-MS/MS and CE-ESI-MS/MS.

Tandem mass spectrometry (MS/MS)-based relative quantification by isobaric labeling is a useful technique to compare different metabolic expression levels in biological systems. For the first time, we have labeled primary and secondary amine-containing small molecules using 4-plex isobaric N,N-dimethyl leucine (DiLeu) to perform relative quantification. Good labeling efficiency and quantification accuracy were demonstrated with a mixture of 12 metabolite standards including amino acids and small molecule neurotransmitters. Labeling amine-containing metabolites with DiLeu reagents also enabled the separation of polar metabolites by nanoRPLC and improved the detection sensitivity by CE-ESI-MS. The 4-plex DiLeu labeling technique combined with LC-MS/MS and CE-MS/MS platforms were applied to profile and quantify amine-containing metabolites in mouse urine. The variability of concentrations of identified metabolites in urine samples from different mouse individuals was illustrated by the ratios of reporter ion intensities acquired from online data-dependent analysis.

Rapid detection of peptide markers for authentication purposes in raw and cooked meat using ambient liquid extraction surface analysis mass spectrometry.

In this Article, our previously developed ambient LESA-MS methodology is implemented to analyze five types of thermally treated meat species, namely, beef, pork, horse, chicken, and turkey meat, to select and identify heat-stable and species-specific peptide markers. In-solution tryptic digests of cooked meats were deposited onto a polymer surface, followed by LESA-MS analysis and evaluation using multivariate data analysis and tandem electrospray MS. The five types of cooked meat were clearly discriminated using principal component analysis and orthogonal partial least-squares discriminant analysis. 23 heat stable peptide markers unique to species and muscle protein were identified following data-dependent tandem LESA-MS analysis. Surface extraction and direct ambient MS analysis of mixtures of cooked meat species was performed for the first time and enabled detection of 10% (w/w) of pork, horse, and turkey meat and 5% (w/w) of chicken meat in beef, using the developed LESA-MS/MS analysis. The study shows, for the first time, that ambient LESA-MS methodology displays specificity sufficient to be implemented effectively for the analysis of processed and complex peptide digests. The proposed approach is much faster and simpler than other measurement tools for meat speciation; it has potential for application in other areas of meat science or food production.

New Approach to Speedup Dynamic Program Parallelization Analysis

Development of parallel programming tools has become a mainstream research topic and it has produced many useful tools that reduce the burden of parallel programming. The most important aspect of any such tool is the data dependency analysis, which is very complex and computationally intensive. Fully automatic tools have many demerits such as low efficiency and lack of interaction with the user. In this paper, the authors present our tool called EasyPar having unique features to assist the user at the time of program development as opposed most of the other tools that work on completed programs. Performing parallelization analysis at the time of program development has a potential to exploit more parallelization opportunity by interaction with the programmer. Two most important requirements of such a tool are the accuracy and the performance of the analysis. The authors propose the method that uses database to enable quick and efficient dependency analysis. The proposed method utilizes a database to save program information in structured from and using a query based method for data dependency analysis. The database approach has three benefits. First, it makes the incremental parsing easier and faster; second, query-based approach makes the dependency analysis efficient and third it allows the demand driven program analysis possible. The authors have tested their tool on popular benchmark codes, presented the performance results, and compared it with relevant work.

Rapid screening of low-molecular-weight phenols from persimmon (Diospyros kaki) pulp using liquid chromatography/UV-visible/electrospray mass spectrometry analysis.

Persimmon fruits have been widely used in traditional medicine owing to their phenolic composition. This research aims to perform a rapid, detailed and affordable study of the profile of low-molecular-weight phenols from persimmon pulp. Two different HPLC-DAD/ESI-MS(n) analyses were performed using a routine three-dimensional ion trap mass spectrometer to analyze the ethanolic extract of persimmon pulp: (1) an untargeted data-dependent analysis to identify the majority of small phenols that included full MS and MS(2) scan events; (2) a targeted data-dependent analysis to identify polymerized phenols (dimers and formic acid adducts) through a source-induced dissociation analysis that included full MS and MS(2) scan events. Thirty-two low-molecular-weight phenols were detected, comprising gallic acid and its glycoside and acyl derivatives, glycosides of p-coumaric, vanillic and cinnamic acids and different flavone di-C-hexosides, most of them reported for the first time in persimmon. The use of a straightforward and affordable methodology of analysis led to obtain an up-to-date profiling of low-molecular-weight phenols in persimmon. The results can help future actions aimed to expand the understanding of the phenolic metabolome of persimmon cultivars.

Mass spectrometric strategies for the identification and characterization of human serum albumin covalently adducted by amoxicillin: ex vivo studies.

This study addresses the detection and characterization of the modification of human serum albumin (HSA) by amoxicillin (AX) in ex vivo samples from healthy subjects under oral amoxicillin administration (acute intake of 1 g every 8 h for 48 h). To reach this goal, we used an analytical strategy based on targeted and untargeted mass spectrometric approaches. Plasma samples withdrawn before AX oral intake represented the negative control samples to test the method selectivity, whereas HSA incubated in vitro with AX was the positive control. Different MS strategies were developed, particularly (1) multiple reaction monitoring (MRM) and precursor ion scan (PIS) using a HPLC system coupled to a triple quadrupole MS analyzer and (2) a dedicated data-dependent scan and a customized targeted MS/MS analysis carried out using a nano-LC system coupled to a high-resolution MS system (LTQ Orbitrap XL). Lys 190 was identified as the only modification site of HSA in the ex vivo samples. The AX adduct was identified and fully characterized by complementary targeted approaches based on triple quadrupole (MRM mode) and orbitrap (SIC mode) mass analyzers. The SIC mode also permitted the relative amount of AX-adducted HSA to be measured, ranging from 1 to 2% (6-12 μM) at 24 and 48 h after the oral intake. No adduct in any ex vivo sample was identified by the untargeted methods (PIS and data-dependent scan mode analysis). The results on one hand indicate that MS, in particular high-resolution MS, analysis represents a suitable analytical tool for the identification/characterization of covalently modified proteins/peptides; on the other hand, they give deeper insight into AX-induced protein haptenation, which is required to better understand the mechanisms involved in AX-elicited allergic reactions.

A loop splitting method for single loops with non-uniform dependences

Computationally expensive programs spend most of their time in the execution of DO-loops [1,2]. Therefore, an efficient approach for exploiting potential parallelism is to concentrate on the parallelism available in loops in ordinary programs and has a considerable effect on the speedup [3,4]. Through some loop transformations using a dependence distance in single loops [6,10], a loop can be splitted into partial loops to be executed in parallel without violating data dependence relations, that is, the size of distance can be used as a reduction factor [6], which is the number of iterations that can be executed in parallel. In the case of non-constant distance such that it varies between different instances of the dependence, it is much more difficult to maximize the degree of parallelism from a loop. Partitioning of loops requires efficient and exact data dependence analysis. A precise dependence analysis helps in identifying dependent/independent iterations of a loop. And it is important that appropriate dependence analysis be applied to exploit maximum parallelism within loops. We can consider some tests that examine the dependence of onedimensional subscripted variables— the separability test, the GCD test and the Banerjee test [8,11]. In general, the GCD test is applied first because of its simplicity, even if it is an approximate test. Next, for the case that the gcd test is true, the separability test is attempted again, and through this exact

Quantitative profiling of bacteriocins present in dairy-free probiotic preparations of Lactobacillus acidophilus by nanoliquid chromatography-tandem mass spectrometry

Bacteriocins are a heterogeneous group of ribosomally synthesized peptides or proteins with antimicrobial activity, produced predominantly by lactic acid bacteria, with potential applications as biopreservatives and probiotics. We describe here a novel strategy based on a bottom-up, shotgun proteomic approach using nanoliquid chromatography-tandem mass spectrometry (nanoLC-MS/MS) with multiple fragmentation techniques for the quantitative profiling of bacteriocins present in the probiotic preparations of Lactobacillus acidophilus. A direct LC-MS/MS analysis with alternate collision-induced dissociation, high-energy collision dissociation, and electron-transfer dissociation fragmentation following a filter-assisted size-exclusion sample prefractionation has resulted in the identification of peptides belonging to 37 bacteriocins or related proteins. Peptides from lactacin F, helveticin J, lysin, avicin A, acidocin M, curvaticin FS47, and carocin D were predominant. The process of freeze drying under vacuum was observed to affect both the diversity and abundance of bacteriocins. Data acquisition using alternating complementary peptide fragmentation modes, especially electron-transfer dissociation, has significantly enhanced the peptide sequence coverage and number of bacteriocin peptides identified. Multi-enzyme proteolytic digestion was observed to increase the sample complexity and dynamic range, lowering the chances of detection of low-abundant bacteriocin peptides by LC-MS/MS. An analytical platform integrating size exclusion prefractionation, nanoLC-MS/MS analysis with multiple fragmentation techniques, and data-dependent decision tree-driven bioinformatic data analysis is novel in bacteriocin research and suitable for the comprehensive bioanalysis of diverse, low-abundant bacteriocins in complex samples.

Multiple Data-Dependent Kernel Fisher Discriminant Analysis for Face Recognition

Kernel Fisher discriminant analysis (KFDA) method has demonstrated its success in extracting facial features for face recognition. Compared to linear techniques, it can better describe the complex and nonlinear variations of face images. However, a single kernel is not always suitable for the applications of face recognition which contain data from multiple, heterogeneous sources, such as face images under huge variations of pose, illumination, and facial expression. To improve the performance of KFDA in face recognition, a novel algorithm named multiple data-dependent kernel Fisher discriminant analysis (MDKFDA) is proposed in this paper. The constructed multiple data-dependent kernel (MDK) is a combination of several base kernels with a data-dependent kernel constraint on their weights. By solving the optimization equation based on Fisher criterion and maximizing the margin criterion, the parameter optimization of data-dependent kernel and multiple base kernels is achieved. Experimental results on the three face databases validate the effectiveness of the proposed algorithm.

Software porting support with component-based and language neutral source code analysis

During their lifetimes, programs are subject to modifications, not only to implement new features or to correct bugs, but also to adapt them to new hardware architectures. Parallel environments as multi/many-cores CPUs or GPUs require source code transformations to extract parallelisable regions of code and to allocate workloads on devices. This process requires not only skilled developers, but mainly a deep comprehension of the semantics of the code. In this work is presented a technique and its prototype implementation for component based and language neutral static program analysis that can assist developers in the analysis, comprehension and transformation of code. Source code is statically analysed and modelled with a language neutral representation conceived to permit interoperability between tool components and plugins. This model is designed as a class hierarchy, and implemented in a library to be used by user-developed add-ons. Parsers for two languages, control flow, data flow, data dependence analysis modules have been implemented or integrated. An algorithmic recogniser has also been integrated: basic algorithmic concepts are extracted in order to feed a knowledge base, upon which an external reasoner can do queries to find instances of known algorithms.

A coverage and slicing dependencies analysis for seeking software security defects.

Software security defects have a serious impact on the software quality and reliability. It is a major hidden danger for the operation of a system that a software system has some security flaws. When the scale of the software increases, its vulnerability has becoming much more difficult to find out. Once these vulnerabilities are exploited, it may lead to great loss. In this situation, the concept of Software Assurance is carried out by some experts. And the automated fault localization technique is a part of the research of Software Assurance. Currently, automated fault localization method includes coverage based fault localization (CBFL) and program slicing. Both of the methods have their own location advantages and defects. In this paper, we have put forward a new method, named Reverse Data Dependence Analysis Model, which integrates the two methods by analyzing the program structure. On this basis, we finally proposed a new automated fault localization method. This method not only is automation lossless but also changes the basic location unit into single sentence, which makes the location effect more accurate. Through several experiments, we proved that our method is more effective. Furthermore, we analyzed the effectiveness among these existing methods and different faults.

Quantitative proteomic dissection of a native 14-3-3ε interacting protein complex associated with hepatocellular carcinoma

The 14-3-3 proteins regulate diverse biological processes that are implicated in cancer development, and seven 14-3-3 isoforms were identified with isoform-specific roles in different human tumors. In our previous work, we dissected the interactome of 14-3-3ε formed during the DNA damage response in a hepatocellular carcinoma (HCC) cell using an AACT/SILAC-based quantitative proteomic approach. In this study, we used a similar proteomic approach to profile/identify the 14-3-3ε interactome formed in native HCC cells. Functional categorization and data-dependent network analysis of the native HCC-specific 14-3-3ε interactome revealed that 14-3-3ε is involved in the regulation of multiple biological processes (BPs)/pathways, including cell cycle control, apoptosis, signal transduction, transport, cell adhesion, carbohydrate metabolism, and nucleic acid metabolism. Biological validation further supports that 14-3-3ε, via association with multiple BP/pathway-specific proteins, coordinates the regulation of proliferation, survival, and metastasis of HCC. The findings in this study, together with those of our previous study, provide an extensive profile of the 14-3-3ε interaction network in HCC cells, which should be valuable for understanding the pathology of HCC and HCC therapy.

Multiplexed and data‐independent tandem mass spectrometry for global proteome profiling

One of the most important early developments in the field of proteomics was the advent of automated data acquisition routines that allowed high-throughput unattended data acquisition during HPLC introduction of peptide mixtures to a tandem mass spectrometer. Prior to this, data acquisition was orders of magnitude less efficient being based entirely on lists of predetermined ions generated in a prior HPLC-MS experiment. This process, known generically as data-dependent analysis, empowered the development of shotgun proteomics where hundreds to thousands of peptide sequences are matched per experiment. In their most popular implementation, the most abundant ionized species from every precursor ion scan at each moment in chromatographic time are successively selected for isolation, activation and tandem mass analysis. While extremely powerful, this strategy has one primary limitation in that detectable dynamic range is restricted (in a top-down manner) to the peptides that ionize the best. To circumvent the serial nature of the data-dependent process and increase detectable dynamic range, the concepts of multiplexed and data-independent acquisition (DIA) have emerged. Multiplexed-data acquisition is based on more efficient co-selection and co-dissociation of multiple precursor ions in parallel, the data from which is subsequently de-convoluted to provide polypeptide sequences for each individual precursor ion. DIA has similar goals, but there is no real-time ion selection based on prior precursor ion scans. Instead, predefined m/z ranges are interrogated either by fragmenting all ions entering the mass spectrometer at every single point in chromatographic time; or by dividing the m/z range into smaller m/z ranges for isolation and fragmentation. These approaches aim to fully utilize the capabilities of mass spectrometers to maximize tandem MS acquisition time and to address the need to expand the detectable dynamic range, lower the limit of detection, and improve the overall confidence of peptide identifications and relative protein quantification measurements. This review covers all aspects of multiplexed- and data-independent tandem mass spectrometry in proteomics, from experimental implementations to advances in software for data interpretation.

Identification of differentially expressed proteins by treatment with PUGNAc in 3T3-L1 adipocytes through analysis of ATP-binding proteome

O-GlcNAc (2-acetamino-2-deoxy-β-D-glucopyranose), an important modification for cellular processes, is catalyzed by O-GlcNAc transferase and O-GlcNAcase. O-(2-acetamido-2-deoxy-D-glucopyranosylidene) amino-N-phenylcarbamate (PUGNAc) is a nonselective inhibitor of O-GlcNAcase, which increases the level of protein O-GlcNAcylation and is known to induce insulin-resistance in adipose cells due to uncharacterized targets of this inhibitor. In this study, using ATP affinity chromatography, we applied a targeted proteomic approach for identification of proteins induced by treatment with PUGNAc. For optimization of proteomic methods using ATP affinity chromatography, comparison of two cell lines (3T3-L1 adipocytes and C2C12 myotubes) and two different digestion steps was performed using four different structures of immobilized ATP-bound resins. Using this approach, based on DNA sequence homologies, we found that the identified proteins covered almost half of ATP-binding protein families classified by PROSITE. The optimized ATP affinity chromatography approach was applied for identification of proteins that were differentially expressed in 3T3-L1 adipocytes following treatment with PUGNAc. For label-free quantitation, a gel-assisted method was used for digestion of the eluted proteins, and analysis was performed using two different MS modes, data-independent (671 proteins identified) and data-dependent (533 proteins identified) analyses. Among identified proteins, 261 proteins belong to nucleotide-binding proteins and we focused on some nucleotide-binding proteins, ubiquitin-activation enzyme 1 (E1), Hsp70, vasolin-containing protein (Vcp), and Hsp90, involved in ubiquitin-proteasome degradation and insulin signaling pathways. In addition, we found that treatment with PUGNAc resulted in increased ubiquitination of proteins in a time-dependent manner, and a decrease in both the amount of Akt and the level of phosphorylation of Akt, a key component in insulin signaling, through downregulation of Hsp90. In this study, based on a targeted proteomic approach using ATP affinity chromatography, we found four proteins related to ubiquitination and insulin signaling pathways that were induced by treatment with PUGNAc. This result would provide insight into understanding functions of PUGNAc in 3T3-L1 cells.

Scalable Parallel Motion Estimation on Muti-GPU System

With NVIDIA’s parallel computing architecture CUDA, using GPU to speed up compute-intensive applications has become a research focus in recent years. In this paper, we proposed a scalable method for multi-GPU system to accelerate motion estimation algorithm, which is the most time consuming process in video encoding. Based on the analysis of data dependency and multi-GPU architecture, a parallel computing model and a communication model are designed. We tested our parallel algorithm and analyzed the performance with 10 standard video sequences in different resolutions using 4 NVIDIA GTX460 GPUs, and calculated the overall speedup. Our results show that a speedup of 36.1 times using 1 GPU and more than 120 times for 4 GPUs on 1920x1080 sequences. Further, our parallel algorithm demonstrated the potential of nearly linear speedup according to the number of GPUs in the system.

Detection and Structural Elucidation of Esterified Oxylipids in Human Synovial Fluid by Electrospray Ionization-Fourier Transform Ion-Cyclotron Mass Spectrometry and Liquid Chromatography-Ion Trap-MS3: Detection of Esterified Hydroxylated Docosapentaenoic Acid Containing Phospholipids

Here, we present the application of a cross-platform approach, combining rapid direct infusion high-resolution/accurate mass electrospray ionization Fourier transform ion-cyclotron mass spectrometry (ESI-FTICRMS) with in-depth data-dependent LC-MS(2) and LC-MS(3) analysis for lipid profiling. The analytical approach as well as the subsequent data handling is described. The method was applied to human synovial fluid samples from osteo- and rheumatoid arthritis patients. Multivariate statistical analysis revealed esterified oxylipids as molecular features in a subset of the patient samples. Employing LC-MS(2) and LC-MS(3) analysis of these species, we were able to clarify the hypothesized lipid structures initially based on the accurate mass measurements performed on the ESI-FTICRMS platform. LC-MS(3) analysis of intact esterified oxy-lipids and LC-MS(2) analysis of the hydrolysis products allowed for the detection of positional isomers. The approach led to the structural elucidation of hydroxylated docosapentaenoic acid-containing diacyl-phosphatidylcholine type phospholipids in human synovial fluid.

An Evaluation Indicator System of Low-carbon Transport for Beijing

Establishing an indicator system is an effective way to find out the exited problems in development low-carbon transport. In this paper, we have analyzed the related principles and established the system of evaluation based on data dependency and principal components analysis (PCA). This system is composed by 23 indicators, taking the core of sustainable development. Based on the index system and weights of index, the paper gives an evaluated level of low-carbon transport in Beijing from 2006 to 2010. The conclusion is that Beijing’s low-carbon transport development level is increasing continuously. Some positive results were achieved form public infrastructure construction, improving ecological environment and upgrading the industrial structure. But there are still some shortcomings, such as decline in bicycle riding, traffic jams and insufficient investment in assets, which reduce the overall level of the low-carbon transport.