1.
|
|
2.
|
|
Using CVMFS on a distributed Kubernetes cluster - The PRP experience
/ Sfiligoi, Igor (speaker) (UCSD)
The Pacific Research Platform (PRP) is operating a Kubernetes cluster that manages over 2.5k CPU cores and 250 GPUs. The compute nodes are distributed over several locations, mostly in California, all connected with high speed networking of 10Gbps or higher.
In order to support OSG users, we were in need of providing CVMFS support in the Kubernetes cluster. [...]
2019 - 1217.
General; CernVM Workshop 2019
External links: Talk details; Event details
In : CernVM Workshop 2019
|
|
3.
|
The pilot way to Grid resources using glideinWMS
/ Sfiligoi, Igor (Fermilab) ; Bradley, Daniel C (Wisconsin U., Madison) ; Holzman, Burt (Fermilab) ; Mhashilkar, Parag (Fermilab) ; Padhi, Sanjay (UC, San Diego) ; Wurthwrin, Frank (UC, San Diego)
Grid computing has become very popular in big and widespread scientific communities with high computing demands, like high energy physics. Computing resources are being distributed over many independent sites with only a thin layer of grid middleware shared between them. [...]
FERMILAB-CONF-09-373-CD.-
5 p.
- Published in : WRI World Congress 2 (2009) 428 - 432
Fulltext: PDF;
|
|
4.
|
CMS experience of running glideinWMS in High Availability mode
/ Sfiligoi, Igor (UC, San Diego) ; Letts, James (UC, San Diego) ; Belforte, Stefano (INFN, Trieste) ; Mc Crea, Alison Jean (UC, San Diego) ; Larson, Krista Elaine (Fermilab) ; Zvada, Marian (Karlsruhe U., EKP) ; Holzman, Burt (Fermilab) ; PÂ Mhashilkar ; Bradley, Daniel Charles (Wisconsin U., Madison) ; Saiz Santos, Maria Dolores (UC, San Diego) et al.
The CMS experiment at the Large Hadron Collider is relying on the HTCondor-based glideinWMS batch system to handle most of its distributed computing needs. In order to minimize the risk of disruptions due to software and hardware problems, and also to simplify the maintenance procedures, CMS has set up its glideinWMS instance to use most of the attainable High Availability (HA) features. [...]
CMS-CR-2013-369.-
Geneva : CERN, 2013 - 6 p.
Fulltext: PDF;
In : 20th International Conference on Computing in High Energy and Nuclear Physics 2013, Amsterdam, Netherlands, 14 - 18 Oct 2013
|
|
5.
|
Using ssh as portal - The CMS CRAB over glideinWMS experience
/ Belforte, Stefano (INFN, Trieste) ; Sfiligoi, Igor (UC, San Diego) ; Letts, James (UC, San Diego) ; Fanzago, Federica (INFN, Padua) ; Saiz Santos, Maria Dolores (UC, San Diego) ; Martin, Terrence (UC, San Diego)
The User Analysis of the CMS experiment is performed in distributed way usingboth Grid and dedicated resources. In order to insulate the users from the details of computing fabric, CMS relies on the CRAB (CMS Remote Analysis Builder) package as an abstraction layer. [...]
CMS-CR-2013-367.-
Geneva : CERN, 2013 - 8 p.
Fulltext: PDF;
In : 20th International Conference on Computing in High Energy and Nuclear Physics 2013, Amsterdam, Netherlands, 14 - 18 Oct 2013
|
|
6.
|
|
7.
|
|
8.
|
|
9.
|
The benefits and challenges of sharing glidein factory operations across nine time zones between OSG and CMS
/ Sfiligoi, Igor (UC, San Diego) ; Dost, Jeffrey Michael (UC, San Diego) ; Zvada, Marian (Karlsruhe U., EKP) ; Butenas, Ignas (Vilnius U.) ; Holzman, Burt (Fermilab) ; Wuerthwein, Frank Karl (UC, San Diego) ; Kreuzer, Peter (RWTH Aachen U.) ; Teige, Scott W. (Indiana U., Bloomington) ; Quick, Robert (Indiana U., Bloomington) ; Hernandez, Jose M. (Madrid, CIEMAT) et al.
OSG has been operating for a few years at UCSD a glideinWMS factory for several scientific communities, including CMS analysis, HCC and GLOW. This setup worked fine, but it had become a single point of failure. [...]
CMS-CR-2012-068.-
Geneva : CERN, 2012 - 7 p.
- Published in : J. Phys.: Conf. Ser. 396 (2012) 032103
Fulltext: PDF;
In : Computing in High Energy and Nuclear Physics 2012, New York, NY, USA, 21 - 25 May 2012, pp.032103
|
|
10.
|
Controlled overflowing of data-intensive jobs from oversubscribed sites
/ Sfiligoi, Igor (UC, San Diego) ; Wuerthwein, Frank Karl (UC, San Diego) ; Bockelman, Brian Paul (Nebraska U.) ; Bradley, Daniel Charles (Wisconsin U., Madison) ; Tadel, Matevz (UC, San Diego) ; Bloom, Kenneth Arthur (Nebraska U.) ; Letts, James (UC, San Diego) ; Mrak Tadel, Alja (UC, San Diego)
The CMS analysis computing model was always relying on jobs running near the data, with data allocation between CMS compute centers organized at management level, based on expected needs of the CMS community. While this model provided high CPU utilization during job run times, there were times when a large fraction of CPUs at certain sites were sitting idle due to lack of demand, all while Terabytes of data were never accessed. [...]
CMS-CR-2012-069.-
Geneva : CERN, 2012 - 9 p.
- Published in : J. Phys.: Conf. Ser. 396 (2012) 032102
Fulltext: PDF;
In : Computing in High Energy and Nuclear Physics 2012, New York, NY, USA, 21 - 25 May 2012, pp.032102
|
|