-
S. Burke,
F. Harris,
I. Stokes-Rees,
I. Augustin,
F. Carminati,
J. Closier,
E. van Herwijnen,
A. Sciaba,
D. Boutigny,
J. J. Blaising, [......],
S. Bagnasco,
P. Cerello,
K. Bos, D. Groep,
W. van Leeuwen,
J. Templon,
O. Smirnova,
O. Maroney,
F. Brochu,
D. Colling
[show abstract]
[hide abstract]
ABSTRACT: An overview is presented of the characteristics of HEP computing and its mapping to the Grid paradigm. This is followed by a synopsis of the main experiences and lessons learned by HEP experiments in their use of DataGrid middleware using both the EDG application testbed and the LCG production service. Particular reference is made to experiment data challenges, and a forward look is given to necessary developments in the framework of the EGEE project.
Journal of Grid Computing 04/2012; 2(4):369-386. · 1.31 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: In large-scale grids with many possible resources (clusters of computing elements) to run applications, it is useful that the resources can provide predictions of job response times so users or resource brokers can make better scheduling decisions. Two metrics need to be estimated for response time predictions: one is how long a job executes on the resource (application run time), the other is how long the job waits in the queue before starting (queue wait time). In this paper we propose an instance based learning technique to predict these two metrics by mining historical workloads. The novelty of our approach is to introduce policy attributes in representing and comparing resource states, which is defined as the pool of running and queued jobs on the resource at the time to make a prediction. The policy attributes reflect the local resource scheduling policies and they can be automatically discovered using a genetic search algorithm. The main advantages of this approach compared with scheduler simulation are two-folds: Firstly, it has a better performance to meet the real time requirement of Grid resource brokering; secondly, it is more general because the scheduling policies are learned from past observations. Our experimental results on the NIKHEF LCG production cluster show that acceptable prediction accuracy can be obtained, where the relative prediction errors for response times are between 0.35 and 0.70.
Grid Computing, 2005. The 6th IEEE/ACM International Workshop on; 12/2005
-
S. A. Morrow,
J. Arneil,
E. C. Aschenauer,
M. F. van Batenburg,
H. P. Blok,
D. J. Boersma,
D. Branford,
T. Davinson,
G. DeMeyer,
J. E. Ducret, [......],
M. Liang,
J. Mackenzie,
C. Marchand,
R. Medaglia,
J. Ryckebusch,
M. van Sambeek,
R. Starink,
G. van der Steenhoven,
M. A. van Uden,
H. de Vries
[show abstract]
[hide abstract]
ABSTRACT: A study of the reaction 12C(e,e'p)11B in parallel kinematics has been made for the missing-momentum region 250<pm<310 MeV/c at incident energies of 379 and 585 MeV. A Rosenbluth separation of the cross sections was carried out to extract the longitudinal and transverse structure functions WL and WT. Calculations that include meson-exchange currents and isobar currents are seen to describe the data better than those using a one-body current only. For the first time, a comparison between the 12C(γ,p) cross section and the transverse part of the 12C(e,e'p) cross section has been made versus an effective momentum that is defined in such a way as to minimize effects due to differences in final-state interactions. This comparison suggests that it may be possible to describe both reactions in one consistent framework if two-body currents are included.
Phys. Rev. C. 01/2005; 71(1).
-
[show abstract]
[hide abstract]
ABSTRACT: In a computational Grid which consists of many computer clusters, job start time predictions are useful to guide resource selections and balance the workload distribution. However, the basic Grid middleware available today either has no means of expressing the time that a site will take before starting a job or uses a simple linear scale. In this paper we introduce a system for predicting job start times on clusters. Our predictions are based on statistical analysis of historical job traces and simulation of site schedulers. We have deployed the system on the EDG (European Data-Grid) production cluster at NIKHEF. The experimental results show that acceptable prediction accuracy is achieved to reflect real site states and site-specific scheduling policies. We find that the average error of our job start time predictions is 18.9 percent of the average job queue wait time and this is around 20 times smaller than the average prediction error using the EDG solution.
Cluster Computing and the Grid, 2004. CCGrid 2004. IEEE International Symposium on; 05/2004
-
W.-J. Kasdorp,
W. H. A. Hesselink, D. Groep,
E. Jans,
N. Kalantar-Nayestanaki,
L. Lapikás,
J. J. van Leeuwe,
A. Misiejuk,
G. J. L. Nooren,
C. J. G. Onderwater,
A. R. Pellegrino,
C. M. Spaltro,
R. Starink,
G. van der Steenhoven,
J. J. M. Steijger,
J. A. Templon
[show abstract]
[hide abstract]
ABSTRACT: The reaction H(e, e
p) has been studied at an invariant mass of 1050 MeV, i.e. well below the (1232) resonance. Cross sections have been obtained at values of , the four-momentum transfer squared, of 0.10, 0.20, and 0.28 (GeV/c), covering a missing-momentum range from 150 to 700 MeV/c. The data are compared to the results of covariant calculations of Tjon, and the results of calculations based on a Schrödinger
formalism due to Laget and the Mainz group, respectively. The data are well described by the calculations of the Mainz group,
whereas they are underestimated by Tjon's calculations at high missing momenta. The calculations of Laget, on the other hand,
overestimate the data at low missing momenta, but give a good account of the data at high missing momenta. More detailed considerations
reveal that the (1232) contributions are dominant at high missing momenta. However, the lacking (1232) contribution in Tjon's calculations is not enough to explain the large discrepancy between his calculation and the
present H(e, ) data at high missing momentum. Probably the deuteron wave function employed in the covariant calculations has a -state contribution that is too small.
Few-Body Systems 01/1998; 25(1):115-132. · 1.44 Impact Factor
-
S Burke,
O Maroney,
F. Brochu,
I Augustin,
F Carminati,
J. Closier,
D Boutigny,
J J Blaising,
V. Garonne,
A. Tsaregorodtsev, [......],
M Reale,
A. De Salvo,
S Bagnasco,
P. Cerello,
O Smirnova,
K Bos,
W van Leeuwen, D. Groep,
F Harris,
I Stokes-Rees
-
LHCb Collaboration,
PR Barbosa-Marinho,
I. Bediaga,
AF Barbosa,
G. Cernicchiaro,
EC de Oliveira,
J. Magnin,
JM de Miranda,
A. Massafferri,
A. Reis, [......],
W Tejessy,
F Teubert,
J Toledo Alarcon,
O Ullaland,
A Valassi,
P. Vazquez Regueiro,
I Videau,
P Wertelaers,
A Wright,
K. Wyllie
-
R. Antunes Nobrega,
A. Franca Barbosa,
I. Bediaga,
G. Cernicchiaro,
E. Correade Oliveira,
J. Magnin,
J. Marquesde Miranda,
A. Massafferri,
E. Polycarpo,
A. Reis, [......],
T Schneider,
A. Schopper,
A Smith,
F Teubert,
N Tuning,
O Ullaland,
P Vannerem,
W Witzeling,
K. Wyllie,
Y Xie
-
R. Antunes Nobrega,
A. Franca Barbosa,
I. Bediaga,
G. Cernicchiaro,
E. Correade Oliveira,
J. Magnin,
J. Marquesde Miranda,
A. Massafferri,
E. Polycarpo,
A. Reis, [......],
T Schneider,
A. Schopper,
A Smith,
F Teubert,
N Tuning,
O Ullaland,
P Vannerem,
W Witzeling,
K. Wyllie,
Y Xie
-
G. De Meyer,
E.C. Aschenauer,
H.P. Blok, D. Groep,
K. Hicks,
H. Holvoet,
E. Jans,
L. Lapikás,
B. Lannoy,
G.J.L. Nooren, [......],
D. Ryckbosch,
J.D. Schipper,
J.J.M. Steijger,
M.F.M. Steenbakkers,
M. Tytgat,
G. van der Steenhoven,
R. Van de Vyver,
L. Van Hoorebeke,
M.A. van Uden,
J. Volmer
[show abstract]
[hide abstract]
ABSTRACT: The knockout of α-clusters induced by electrons has been used to study four-nucleon correlations in 16O. For the transition to the ground state of 12C the (e,e′α) cross section has been measured for three values of the momentum transfer and four values of the missing momentum. The momentum-transfer dependence gives evidence for a quasi-elastic α-knockout mechanism. The shape of the measured missing-momentum distribution is reasonably well described by shell-model and cluster-model calculations, but the theoretical curves overshoot the data by a factor three to four.
Physics Letters B.
-
R. Sturrock,
R. Bischof,
B Epp,
V M Ghete,
D Kuhn,
A.G. Mello,
B Caron,
M.C. Vetterli,
G Karapetian,
K. Martens, [......],
F. Luehring,
S Goldfarb,
H Severini,
P Skubic,
Y Gao,
T Ryan,
K De,
M Sosebee,
P. McGuigan,
N. Ozturk
[show abstract]
[hide abstract]
ABSTRACT: The ATLAS Collaboration at CERN is preparing for the data taking and analysis at the LHC that will start in 2007. Therefore, a series of Data Challenges was started in 2002 whose goals are the validation of the Computing Model, of the complete software suite, of the data model, and to ensure the correctness of the technical choices to be made for the final offline computing environment. A major feature of the first Data Challenge (DC1) was the preparation and the deployment of the software required for the production of large event samples as a worldwide distributed activity. It should be noted that it was not an option to "run the complete production at CERN" even if we had wanted to; the resources were not available at CERN to carry out the production on a reasonable time-scale. The great challenge of organising and carrying out this large-scale production at a significant number of sites around the world had therefore to be faced. However, the benefits of this are manifold: apart from realising the required computing resources, this exercise created worldwide momentum for ATLAS computing as a whole. This report describes in detail the main steps carried out in DC1 and what has been learned form them as a step towards a computing Grid for the LHC experiments.
-
R. Sturrock,
R. Bischof,
B Epp,
V M Ghete,
D Kuhn,
A.G. Mello,
B Caron,
M.C. Vetterli,
G V Karapetian,
K. Martens, [......],
F C Luehring,
S Goldfarb,
H Severini,
P L Skubic,
Y Gao,
T Ryan,
K De,
M Sosebee,
P. McGuigan,
N. Ozturk
[show abstract]
[hide abstract]
ABSTRACT: The ATLAS Collaboration at CERN is preparing for the data taking and analysis at the LHC that will start in 2007. Therefore, a series of Data Challenges was started in 2002 whose goals are the validation of the Computing Model, of the complete software suite, of the data model, and to ensure the correctness of the technical choices to be made for the final offline computing environment. A major feature of the first Data Challenge (DC1) was the preparation and the deployment of the software required for the production of large event samples as a worldwide distributed activity. It should be noted that it was not an option to "run the complete production at CERN" even if we had wanted to; the resources were not available at CERN to carry out the production on a reasonable time-scale. The great challenge of organising and carrying out this large-scale production at a significant number of sites around the world had therefore to be faced. However, the benefits of this are manifold: apart from realising the required computing resources, this exercise created worldwide momentum for ATLAS computing as a whole. This report describes in detail the main steps carried out in DC1 and what has been learned form them as a step towards a computing Grid for the LHC experiments.
