The Experts below are selected from a list of 5820 Experts worldwide ranked by ideXlab platform
Vandelli Waine - One of the best experts on this subject based on the ideXlab platform.
-
Automated load balancing in the ATLAS high-performance storage software
'Springer Science and Business Media LLC', 2017Co-Authors: Le Goff Fabrice, Vandelli WaineAbstract:The ATLAS experiment collects proton-proton collision events delivered by the LHC accelerator at CERN. The ATLAS Trigger and Data Acquisition (TDAQ) system selects, transports and eventually records event data from the detector at several gigabytes per second. The data are recorded on transient storage before being delivered to permanent storage. The transient storage consists of high-performance direct-attached storage servers accounting for about 500 hard drives. The transient storage operates dedicated software in the form of a distributed multi-threaded application. The workload includes both CPU-demanding and IO-oriented tasks. This paper presents the original application Threading Model for this particular workload, discussing the load-sharing strategy among the available CPU cores. The limitations of this strategy were reached in 2016 due to changes in the trigger configuration involving a new data distribution pattern. We then describe a novel data-driven load-sharing strategy, designed to automatically adapt to evolving operational conditions, as driven by the detector configuration or the physics research goals. The improved efficiency and adaptability of the solution were measured with dedicated studies on both test and production systems. This paper reports on the results of those tests which demonstrate the capability of operating in a large variety of conditions with minimal user intervention.ATLAS [1] is one of the general purpose detectors observing proton-proton collisions provided by the LHC [2] at CERN. The ATLAS Trigger and Data Acquisition (TDAQ) system [3] is responsible for conveying the event data from the detector up to a permanent mass-storage system provided by CERN. This work focuses on the Data Logger system which lies at the end of the data flow path in the TDAQ system. The Data Logger is a transient storage system recording the selected event data on hard drives before transferring them to permanent storage where they are available for offline analysis
-
Automated load balancing in the ATLAS high-performance storage software
2017Co-Authors: Le Goff Fabrice, Vandelli WaineAbstract:The ATLAS experiment collects proton-proton collision events delivered by the LHC accelerator at CERN. The ATLAS Trigger and Data Acquisition (TDAQ) system selects, transports and eventually records event data from the detector at several gigabytes per second. The data are recorded on transient storage before being delivered to permanent storage. The transient storage consists of high-performance direct-attached storage servers accounting for about 500 hard drives. The transient storage operates dedicated software in the form of a distributed multi-threaded application. The workload includes both CPU-demanding and IO-oriented tasks. This paper presents the original application Threading Model for this particular workload, discussing the load-sharing strategy among the available CPU cores. The limitations of this strategy were reached in 2016 due to changes in the trigger configuration involving a new data distribution pattern. We then describe a novel data-driven load-sharing strategy, designed to automatically adapt to evolving operational conditions, as driven by the detector configuration or the physics research goals. The improved efficiency and adaptability of the solution were measured with dedicated studies on both test and production systems. This paper reports on the results of those tests which demonstrate the capability of operating in a large variety of conditions with minimal user intervention
Bernd H A Rehm - One of the best experts on this subject based on the ideXlab platform.
-
replacement of the catalytic nucleophile cysteine 296 by serine in class ii polyhydroxyalkanoate synthase from pseudomonas aeruginosa mediated synthesis of a new polyester identification of catalytic residues
Biochemical Journal, 2003Co-Authors: Amro A Amara, Bernd H A RehmAbstract:The class II PHA (polyhydroxyalkanoate) synthases [PHA(MCL) synthases (medium-chain-length PHA synthases)] are mainly found in pseudomonads and catalyse synthesis of PHA(MCL)s using CoA thioesters of medium-chain-length 3-hydroxy fatty acids (C6-C14) as a substrate. Only recently PHA(MCL) synthases from Pseudomonas oleovorans and Pseudomonas aeruginosa were purified and in vitro activity was achieved. A Threading Model of the P. aeruginosa PHA(MCL) synthase PhaC1 was developed based on the homology to the epoxide hydrolase (1ek1) from mouse which belongs to the alpha/beta-hydrolase superfamily. The putative catalytic residues Cys-296, Asp-452, His-453 and His-480 were replaced by site-specific mutagenesis. In contrast to class I and III PHA synthases, the replacement of His-480, which aligns with the conserved base catalyst of the alpha/beta-hydrolases, with Gln did not affect in vivo enzyme activity and only slightly in vitro enzyme activity. The second conserved histidine His-453 was then replaced by Gln, and the modified enzyme showed only 24% of wild-type in vivo activity, which indicated that His-453 might functionally replace His-480 in class II PHA synthases. Replacement of the postulated catalytic nucleophile Cys-296 by Ser only reduced in vivo enzyme activity to 30% of wild-type enzyme activity and drastically changed substrate specificity. Moreover, the C296S mutation turned the enzyme sensitive towards PMSF inhibition. The replacement of Asp-452 by Asn, which is supposed to be required as general base catalyst for elongation reaction, did abolish enzyme activity as was found for the respective amino acid residue of class I and III enzymes. In the Threading Model residues Cys-296, Asp-452, His-453 and His-480 reside in the core structure with the putative catalytic nucleophile Cys-296 localized at the highly conserved gamma-turns of the alpha/beta-hydrolases. Inhibitor studies indicated that catalytic histidines reside in the active site. The conserved residue Trp-398 was replaced by Phe and Ala, respectively, which caused inactivation of the enzyme indicating an essential role of this residue. In the Threading Model this residue was found to be surface-exposed. No evidence for post-translational modification by 4-phosphopantetheine was obtained. Overall, these data suggested that in class II PHA synthases the conserved histidine which was found as general base catalyst in the catalytic triad of enzymes related to the alpha/beta-hydrolase superfamily, was functionally replaced by His-453 which is conserved among all PHA synthases.
-
molecular characterization of the poly 3 hydroxybutyrate phb synthase from ralstonia eutropha in vitro evolution site specific mutagenesis and development of a phb synthase protein Model
Biochimica et Biophysica Acta, 2002Co-Authors: Bernd H A Rehm, Amro A Amara, Regina V Antonio, Patricia Spiekermann, Alexander SteinbuchelAbstract:A Threading Model of the Ralstonia eutropha polyhydroxyalkanoate (PHA) synthase was developed based on the homology to the Burkholderia glumae lipase, whose structure has been resolved by X-ray analysis. The lid-like structure in the Model was discussed. In this study, various R. eutropha PHA synthase mutants were generated employing random as well as site-specific mutagenesis. Four permissive mutants (double and triple mutations) were obtained from single gene shuffling, which showed reduced activity and whose mutation sites mapped at variable surface-exposed positions. Six site-specific mutations were generated in order to identify amino acid residues which might be involved in substrate specificity. Replacement of residues T323 (I/S) and C438 (G), respectively, which are located in the core structure of the PHA synthase Model, abolished PHA synthase activity. Replacement of the two amino acid residues Y445 (F) and L446 (K), respectively, which are located at the surface of the protein Model and adjacent to W425, resulted in reduced activity without changing substrate specificity and indicating a functional role of these residues. The E267K mutant exhibited only slightly reduced activity with a surface-exposed mutation site. Four site-specific deletions were generated to evaluate the role of the C-terminus and variant amino acid sequence regions, which link highly conserved regions. Deleted regions were D281-D290, A372-C382, E578-A589 and V585-A589 and the respective PHA synthases showed no detectable activity, indicating an essential role of the variable C-terminus and the linking regions between conserved blocks 2 and 3 as well as 3 and 4. Moreover, the N-terminal part of the class II PHA synthase (PhaC(Pa)) from Pseudomonas aeruginosa and the C-terminal part of the class I PHA synthase (PhaC(Re)) from R. eutropha were fused, respectively, resulting in three fusion proteins with no detectable in vivo activity. However, the fusion protein F1 (PhaC(Pa)-1-265-PhaC(Re)-289-589) showed 13% of wild type in vitro activity with the fusion point located at a surface-exposed loop region.
Alexander Steinbuchel - One of the best experts on this subject based on the ideXlab platform.
-
molecular characterization of the poly 3 hydroxybutyrate phb synthase from ralstonia eutropha in vitro evolution site specific mutagenesis and development of a phb synthase protein Model
Biochimica et Biophysica Acta, 2002Co-Authors: Bernd H A Rehm, Amro A Amara, Regina V Antonio, Patricia Spiekermann, Alexander SteinbuchelAbstract:A Threading Model of the Ralstonia eutropha polyhydroxyalkanoate (PHA) synthase was developed based on the homology to the Burkholderia glumae lipase, whose structure has been resolved by X-ray analysis. The lid-like structure in the Model was discussed. In this study, various R. eutropha PHA synthase mutants were generated employing random as well as site-specific mutagenesis. Four permissive mutants (double and triple mutations) were obtained from single gene shuffling, which showed reduced activity and whose mutation sites mapped at variable surface-exposed positions. Six site-specific mutations were generated in order to identify amino acid residues which might be involved in substrate specificity. Replacement of residues T323 (I/S) and C438 (G), respectively, which are located in the core structure of the PHA synthase Model, abolished PHA synthase activity. Replacement of the two amino acid residues Y445 (F) and L446 (K), respectively, which are located at the surface of the protein Model and adjacent to W425, resulted in reduced activity without changing substrate specificity and indicating a functional role of these residues. The E267K mutant exhibited only slightly reduced activity with a surface-exposed mutation site. Four site-specific deletions were generated to evaluate the role of the C-terminus and variant amino acid sequence regions, which link highly conserved regions. Deleted regions were D281-D290, A372-C382, E578-A589 and V585-A589 and the respective PHA synthases showed no detectable activity, indicating an essential role of the variable C-terminus and the linking regions between conserved blocks 2 and 3 as well as 3 and 4. Moreover, the N-terminal part of the class II PHA synthase (PhaC(Pa)) from Pseudomonas aeruginosa and the C-terminal part of the class I PHA synthase (PhaC(Re)) from R. eutropha were fused, respectively, resulting in three fusion proteins with no detectable in vivo activity. However, the fusion protein F1 (PhaC(Pa)-1-265-PhaC(Re)-289-589) showed 13% of wild type in vitro activity with the fusion point located at a surface-exposed loop region.
Rio Yokota - One of the best experts on this subject based on the ideXlab platform.
-
an fmm based on dual tree traversal for many core architectures
Journal of Algorithms & Computational Technology, 2013Co-Authors: Rio YokotaAbstract:The present work attempts to integrate the independent efforts in the fast N-body community to create the fastest N-body library for many-core and heterogenous architectures. Focus is placed on low accuracy optimizations, in response to the recent interest to use FMM as a preconditioner for sparse linear solvers. A direct comparison with other state-of-the-art fast N-body codes demonstrates that orders of magnitude increase in performance can be achieved by careful selection of the optimal algorithm and low-level optimization of the code. The current N-body solver uses a fast multipole method with an efficient strategy for finding the list of cell-cell interactions by a dual tree traversal. A task-based Threading Model is used to maximize thread-level parallelism and intra-node load-balancing. In order to extract the full potential of the SIMD units on the latest CPUs, the inner kernels are optimized using AVX instructions.
-
an fmm based on dual tree traversal for many core architectures
arXiv: Numerical Analysis, 2012Co-Authors: Rio YokotaAbstract:The present work attempts to integrate the independent efforts in the fast N-body community to create the fastest N-body library for many-core and heterogenous architectures. Focus is placed on low accuracy optimizations, in response to the recent interest to use FMM as a preconditioner for sparse linear solvers. A direct comparison with other state-of-the-art fast N-body codes demonstrates that orders of magnitude increase in performance can be achieved by careful selection of the optimal algorithm and low-level optimization of the code. The current N-body solver uses a fast multipole method with an efficient strategy for finding the list of cell-cell interactions by a dual tree traversal. A task-based Threading Model is used to maximize thread-level parallelism and intra-node load-balancing. In order to extract the full potential of the SIMD units on the latest CPUs, the inner kernels are optimized using AVX instructions. Our code -- exaFMM -- is an order of magnitude faster than the current state-of-the-art FMM codes, which are themselves an order of magnitude faster than the average FMM code.
Aswani Kumar Singh - One of the best experts on this subject based on the ideXlab platform.
-
improving rsa algorithm using multi Threading Model for outsourced data security in cloud storage
International Conference on Cloud Computing, 2018Co-Authors: Purnima Gupta, Deepak Kumar Verma, Aswani Kumar SinghAbstract:Cloud Computing is a promising technology used by scientific and enterprise communities to access shared resources from anywhere through the internet. Users may also store their sensitive and confidential information on the cloud that requires an eminent encryption scheme and a fine-grained access policy to ensure privacy and security. Disparate obtainable symmetric, asymmetric and Attribute-Based Encryption has appreciable encryption schemes to keep secure the confidential data. In this paper, the authors have analyzed existing data encryption schemes, like RSA, KP-ABE, CP-ABE, and AES. The comparisons among them on the basis of computational cost and storage cost have been shown. Further, the authors have proposed an improving scheme to enhance the speed of RSA encryption using multi-Threading concept on latest multi-core CPUs.
