Mr
Pierrick Micout
(CEA DAPNIA Saclay)
05/11/2007, 09:30
Site Reports
What is new since the last HEPiX fall meeting in DAPNIA Saclay.
Mr
Michel Jouvin
(LAL / IN2P3)
05/11/2007, 10:10
Site Reports
Site report about LAL and GRIF (LCG T2)
Dr
Roberto Gomezel
(INFN)
05/11/2007, 11:40
Site Reports
an overall report about updates of INFN computing infrustructure
Manfred Alef
(Forschungszentrum Karlsruhe)
05/11/2007, 12:00
Site Reports
Brief presentation of the current status at the Grid Computing Centre Karlsruhe.
Mr
Rafal Otto
(CERN)
05/11/2007, 14:00
Windows
Windows Vista has been released by Microsoft early this year and Office 2007 even at the end of the last year.
First, in March 2007, Office 2007 pilot has been announced at CERN. This new Office suite version is now pre-installed on each new computer at CERN. In addition users can upgrade to Office 2007 their existing PCs. With such deployment schema we currently have Office 2007 installed...
Mr
Wayne Baisley
(Fermilab)
05/11/2007, 14:30
Windows
Experiences with Microsoft Key Management Server
Kazimierz Popinski
(DESY)
05/11/2007, 15:30
Windows
Group Policy Objects (GPOs) have a pole position in a domain network management.
Unfortunately, the primary tool for managing Group Policy in an enterprise, the Group Policy Management Console (GPMC), doesn’t provide a base for a workflow in any but the simplest management models.
Microsoft Advanced Group Policy Management (AGPM), the technology included in the Microsoft Desktop...
Mr
Alexandre Lossent
(CERN)
05/11/2007, 16:00
General
CERN recently introduced a new type of web sites labelled "Collaboration Workspaces", based on SharePoint 2007. It aims at facilitating a number of tasks like web publishing, information gathering, and collaborative work on documents.
We will present the use cases of SharePoint at CERN, brief technical details about the deployment and how we integrated SharePoint with existing web sites and...
Dr
Walter Schoen
(GSI)
05/11/2007, 16:30
Storage
Evaluation of a clustre file system for the GSI data file system
- architecture of the test setup
- RAID controller/disk array performance
- OST I/O performance
- client I/O perfomance
- clustre file system performance
- trunking/bonding
Wolfgang Friebel
(DESY)
06/11/2007, 09:00
Site Reports
DESY site report - Update on recent activities at DESY
Mattias Wadenstein
(NDGF)
06/11/2007, 09:20
Site Reports
Current status at the distributed tier1 "site" NDGF.
Dr
Iwona Sakrejda
(LBNL/NERSC)
06/11/2007, 09:40
Site Reports
Recent changes in PDSF organization. Benefits and pitfalls of a shared resource.
Sandy Philpott
(JLab)
06/11/2007, 10:00
Site Reports
Update on JLab computing since the last HEPiX meeting.
Gary Stiehr
(The Genome Sequencing Center at Washington University in St. Louis)
06/11/2007, 11:50
Site Reports
Information about the current IT infrastructure at the Genome Sequencing Center at Washington University in St. Louis and specifically how it has changed over the last year.
Dr
Guy Coates
(The Wellcome Trust Sanger Institute)
06/11/2007, 12:10
Site Reports
Sanger Institute site report
Andrei Maslennikov
(CASPUR)
06/11/2007, 13:35
Storage
The talk will be covering the work done during the second
phase of HEPIX/IHEPCCC File Systems /Storage Working Group.
The plans for the third and final phase will also be
presented.
Dr
Iwona SAKREJDA
(LBNL/NERSC)
06/11/2007, 14:00
Storage
NERSC storage update. What's happening and where are we going.
Mr
Jacques-Charles Lafoucriere
(CEA)
06/11/2007, 14:30
Storage
Lustre is a high performance parallel filesystem. It manages
Petabytes of storage and allows hundreds of GB/s. Lustre HSM
project will bring to Lustre ILM functionalities. This talk
will describe the design of Lustre HSM.
John Josephakis
(Data Direct Networks)
06/11/2007, 15:30
Storage
Storage devices with the native Infiniband interface are gaining
more and more popularity. I will discuss the available Infiniband
storage access protocols, and will then use one of our recent
products as an example.
Mr
Peter van der Reest
(DESY - IT)
06/11/2007, 16:00
Storage
With the start of the Large Hadron Collider, presumably in 2008, the largest share of the produced and generated physics data outside CERN will be managed by the dCache storage element. Beside the challenge to handle storage in the order of tens of Petabyte in a single installation, the number of file operations per second is certainly even more challenging. Two years ago we identified PNFS,...
Derrick Brashear
(Sine Nomine Associates and OpenAFS Gatekeeper),
Jeffrey Altman
(Secure Endpoints Inc. and OpenAFS Gatekeeper)
06/11/2007, 16:30
Storage
OpenAFS has completed seven years as an open source project.
During that time OpenAFS has successfully tackled many challenges
including new platforms, multi-processor systems, and network
environments consisting of mobile clients and partitioning.
OpenAFS Gatekeepers, Derrick Brashear and Jeffrey Altman, will
provide updates on the latest OpenAFS offerings, the most common
usage models,...
Dr
Tony Cass
(CERN)
07/11/2007, 09:00
Data Centers
I will give a general introduction to the problems faced by today's data centres and summarise the responses to the survey of HEPiX sites carried out prior to this meeting.
Dr
Antonio Chan
(Brookhaven National Lab)
07/11/2007, 09:30
Data Centers
Brookhaven is the primary computing center for RHIC, and it is also a Tier 1 computing center
for the ATLAS experiment at CERN. The growing computing needs from both RHIC and
ATLAS has exceeded the capacity of the current computer center, and a long-term expansion
plan is now underway to meet future requirements. This presentation discusses our expansion
plans as well as possible...
Gary Stiehr
(The Genome Sequencing Center at Washington University in St. Louis)
07/11/2007, 10:00
Data Centers
The Washington University School of Medicine is building a new data
center with the needs of the Genome Sequencing Center in mind. It is
being constructed to have full power and cooling redundancy for 120
racks of equipment, with an anticipated mixture of 2/3 8 kW storage
racks and 1/3 25 kW compute racks. We will discuss its design, the
problems we encountered along the way, our current...
Mr
Martin Bly
(STFC - RAL)
07/11/2007, 11:30
Data Centers
We describe the processes surrounding the procurement of a new machine room for the RAL site.
Mr
Charles Boeheim
(SLAC)
07/11/2007, 12:00
Data Centers
SLAC recently acquired a Sun Project Black Box for expansion of its compute clusters. I will describe our experiences with installing and operating the box and its contents, and go over some of the economic tradeoffs in expansion strategies.
Mrs
Veronique Lefebure
(CERN Staff)
07/11/2007, 13:30
Virtualization
Mr
Alex Iribarren
(CERN)
07/11/2007, 13:40
Mr
Federico Nebiolo
(INFN)
07/11/2007, 14:10
Virtualization
Virtualization can enhance the functionality and ease the management of current and future Grids by enabling on-demand creation of services and virtual clusters with customized environments, dynamic provisioning and policy-based resource allocation, as well as enhance high availability and load balancing techniques.
In this work, we consider the work done in the last year in different INFN...
Mr
Oliver Oberst
(FZK)
07/11/2007, 14:30
Jan Svec
(CESNET)
07/11/2007, 15:30
Virtualization
New scientific communities, attracted by successful deployment of Grids by
various scientific groups, bring new requirements on Grid middleware, such as
fast turn-around for short jobs, access prioritisation, and need for various
execution environment optimized to application specific requirements.
We will argue that virtualized Grids, based on idea that different virtual
worker nodes,...
Dr
Keith Chadwick
(Fermilab)
07/11/2007, 15:50
Virtualization
We will discuss the deployment of Linux Virtualisation within the Fermilab Campus Grid (FermiGrid) infrastructure to provide highly available Grid services.
Ian Gable
(University of Victoria)
07/11/2007, 16:50
Virtualization
HEP applications often have very specific OS requirements and shared
resources available to HEP in Canada may not be able to meet the
specific OS requirements of a particular application. We have
investigated VMs as a mechanism for packaging a complete application and
shipping it out to a remote grid site complete with all its requirements.
We will discuss experiences using Globus Virtual...
Dr
Keith Chadwick
(Fermilab)
08/11/2007, 09:00
Grid Management and Monitoring
We will present the tools which are used to monitor the grid services within the Fermilab Campus Grid (FermiGrid).
James Casey
(CERN)
08/11/2007, 09:30
Grid Management and Monitoring
We will demonstrate two new tools for monitoring grid services within WLCG. The first is a set of grid probes, designed to test EGEE and OSG services at a site, integrated into nagios. The second is a new visualisation tool - GridMap (http://gridmap.cern.ch). This uses heatmaps to compactly display site availability for the entire WLCG Grid on a single page.
Dr
Benjamin Gaidioz
(CERN)
08/11/2007, 10:00
Grid Management and Monitoring
The presentation will give an overview of the Experiment Dashboard monitoring applications, providing LHC users with the possibility to monitor analysis and production activities on the distributed infrastructure. Experiment Dashboard monitoring applications are working in a transparent way across several middleware platforms used by the LHC experiments (LCG/gLite,OSG,NorduGrid).
The...
Dr
David Kelsey
(RAL)
08/11/2007, 11:00
Grid Management and Monitoring
This talk will describe recent progress on Grid Security in WLCG and EGEE covering technical, operational and policy issues.
Iwona Sakrejda
(LBNL)
08/11/2007, 11:30
Grid Management and Monitoring
The DOE SciDAC Center for Enabling Distributed Petascale Science
(CEDPS) has been working with the Open Science Grid to
design and deploy tools to facilitate troubleshooting problems on the
Grid. This includes tools for central log collection, and advice on
what should be logged. This talk will give an overview of progress to
date and near-term plans for the future.
Dr
Helge Meinhard
(CERN-IT)
08/11/2007, 12:00
Benchmarking
Just a few words to get the session kicked off...
Dr
michele michelotto
(INFN Padova)
08/11/2007, 12:05
Benchmarking
INFN Computing Committee is looking for a modern benchmark to replace the
SPEC Int 2000 for Worker Node evaluation. Performance results from several HEP experiments have been compared with SPEC Int 2000 and Int 2006 to find the best agreement.
Manfred Alef
(Forschungszentrum Karlsruhe)
08/11/2007, 13:30
Benchmarking
I'll continue with my series of talks about CPU benchmarking issues.
New topics are:
- SPECint2006 measurements, what's the difference to SPECint2000?
- SPECfp2006 measurements.
Dr
Helge Meinhard
(CERN-IT)
08/11/2007, 14:00
Benchmarking
A collection of remarks on recent procurements and our experience with including power consumption, SPECpower etc.
Dr
Don Holmgren
(Fermilab)
08/11/2007, 14:30
Benchmarking
We will present lattice QCD applications performance data on current Intel and AMD processors and platforms, including the new quad-core AMD processors ("Barcelona").
Jim Williams
(Oracle Corp.)
08/11/2007, 15:30
Storage
The loss of a business critical application resulting from data
corruption can be crippling and require huge efforts to repair.
All businesses hope they have the know how to recover, yet no
business wants an unplanned real-time test. Data corruption
prevention, rather than recovery is key. Storage vendors,
including those designing adapters, fabrics, and storage arrays
design...
Mr
Andras Horvath
(CERN)
08/11/2007, 16:30
Data Centers
In the recent years the Intelligent Platform Management Interface (IPMI)
has matured and become widespread enough that even commodity hardware
can now be reliably and securely managed using its functionality.
We report on the successful large-scale deployment of secure remote
power control of machines and the integration of IPMI Serial-Over-LAN
functionality into the existing remote...
Troy Dawson
(FNAL)
09/11/2007, 09:00
General
Progress of Scientific Linux over the past 6 months.
What we are currently working on.
What we see in the future for Scientific Linux
Troy Dawson
(FNAL)
09/11/2007, 09:30
General
Feedback to and input for the SL developers from the HEPiX community. This may influence upcoming decisions e.g. on distribution lifecycles.
Mr
Charles Boeheim
(SLAC)
09/11/2007, 10:00
General
"GUIs don't scale" has long been one of my mantras. I will demonstrate some recent work I have done in trying to overturn this limitation. By using very high density display techniques such as treemapping and pixel charts, I have been able to create some displays that show the status and history of thousands of machines without abstracting all of the meaningful detail out of them.
Mr
Alan Silverman (Speaker)
(CERN)
09/11/2007, 11:00
General
Integrated Site Security for Grids (ISSeG) is a project co-funded by the
European Commission involving CERN, STFC and FZK. Integrated Site
Security (ISS) coordinates improvements so that policies, rules,
awareness and training elvolve in step with technological developments.
This session will cover the resources available on www.isseg.eu:
- Risk assessment questionnaire
- Security...
Mr
Rafal Otto
(CERN)
09/11/2007, 11:30
General
Nearly every large organization uses a tool to broadcast messages and information across the internal campus (messages like alerts announcing interruption in services or just information about upcoming events). These tools typically allow administrators (operators) to send "targeted" messages which are sent only to specific groups of users or computers, e/g only those located in a specified...
Mr
Alexandre Lossent
(CERN)
09/11/2007, 12:00
General
In order to streamline access to applications, CERN recently deployed Single-Sign-On (SSO) mecanisms along with changes in identity management and access management.
We will present the selected SSO solution and how it integrates with existing applications, as well as related changes in user accounts and roles.