Posts

UPDATED: Cygnus/Force: Second failure of new VM Storage (FIXED)

————————————————————

UPDATE: At 8:45pm EDT, Force resumed normal function. The normal computing environment is now restored.

————————————————————

UPDATE: At 7:35pm EDT, Cygnus resumed normal function. Force is still under repair.

————————————————————

5:30pm:

Well folks, I hate to do this to you again, but it looks like I need
to take cygnus and force down again thanks to problems with the storage.

Again, I’ll down Cygnus & Force at 7pm EDT. Please begin the process
of saving your work.

At this point, I’m moving these back to the old storage system, which,
while slow (and it did impact the responsiveness of these machines) at
least stayed running without issues. The new machine has not
previously shown issues in its prior use, so, I admit to being a bit
flummoxed as to what is going on.

This downtime will be longer as I need to scrub a few things clean,
make sure the VMs will be intact and usable.

I’ll let you know when things are back online. I don’t have good
estimates this time.

No scheduled compute jobs will be impacted.

I, and the rest of the PACE team apologize for the continued
interruption in service and we hope to rectify these issues in a
couple of hours from now.

Thanks for your patience.

bnm

Urgent: Cygnus & FoRCE head nodes reboot at 7pm due to Storage issues

Hey folks,

We suffered a temporary loss of connectivity to the backend storage
serving our VM farm earlier this afternoon. As such, several running
VMs moved their OS filesystems to a read-only state.

The filesystems on which your data is stored are fine, however.

Unfortunately, though, the head nodes for Cygnus and the FoRCE
clusters were affected, and judging by our previous experience with
this, we need to reboot these nodes soon. As such, we ask any
currently logged in users to please save their data now and logout.

We are scheduling a reboot of these systems at 7:00pm EDT. A few
minutes after that, the nodes should be available and fully functional.

No jobs have been, nor will be lost in this process.

We are sorry for the inconvenience, and plan to keep you up to date
with any further issues with these, as well as the rest of the machines.

We’re back up

The maintenance day ran rather a bit longer than anticipated but the clusters are now back in operation and processing jobs. As usual, please send any reports of trouble to pace-support@oit.gatech.edu.

Upcoming quarterly maintenance – 10/18/2011

Reminder folks, the clusters will be down on this coming Tuesday, October 18.

All of the currently running jobs will have completed by then, and the scheduler has been instructed to not start any new jobs that will not complete by then. Jobs that have been submitted, but wouldn’t complete by Tuesday morning are being held by the scheduler, and will be released as nodes become available after our maintenance activities.

Major items on the list this time around are:

  • swap over to redundant network switches for the core of the HPC network
  • Panasas software update to version 4.1
  • routine Solaris and RedHat patching to non-user facing infrastructure services
  • routine security patches to ssh everywhere
  • migration of infrastructure services to virtual machines
  • migration to new infrastructure-facing LDAP schema
  • reinstating storage quotas missed in our previous maintenance

Some further minor things we’ll take care of as well:

  • load testing on some infrastructure servers
  • migrate the /hp3 filesystem to different fileserver, we put it on the wrong one; (no user impact expected)
  • OIT/Operations will be performing preventative maintenance on the UPS
  • OIT/Operations will be verifying some electrical circuit locations
  • update ganglia monitoring agents on all RHEL5 machines
  • reboot everything

 

Technical Computing with MATLAB at Emory University

MathWorks is running a seminar over at Emory.  Please see http://www.mathworks.com/company/events/seminars/seminar60145.html for the details.  This technical seminar will discuss:

  • Importing data from Excel in to MATLAB
  • Using powerful graphics features to quickly and easily visualize and understand your data
  • Processing data using pre-written functions
  • Dynamically link MATLAB to Excel
  • Share your result
  • Speed up MATLAB applications with Parallel Computing
  • Handling large datasets with Parallel Computing

 

PACE passes 10,000 CPU core milestone

As we are making progress with the installation of our latest round of faculty purchases, we have now passed a significant milestone – 10,000 CPU cores managed by the PACE team.  Woohoo!  Along with all of those processors, we are also managing over half a petabyte of storage across the faculty project space, infrastructure and backups, plus a couple more hundred terabytes on our high-performance filesystems.

Given the amount of spending we have in progress at the moment, it’s very likely that we will surpass the 15,000 CPU core mark by the end of the calendar 2011.  We’ll also likely get close to a petabyte of storage.

 

Wow.

 

I’d like to take a moment and thank all of you for placing your trust in us.  On behalf of myself and the entire PACE team, I look forward to helping you take Georgia Tech to even higher levels in the future.

 

THANK YOU!!

MADE@GT OCT. 14

Friends,

I invite you to attend a technical seminar: “MAterials DEsign @ Georgia Tech” (MADE@GT) on Friday Oct. 14, 2011 @ the Georgia Tech Hotel & Conference Center, from 9A-5P.  The complimentary event also includes free continental breakfast, lunch, and closing reception.

The MADE@GT seminar (agenda below) consists of 10 technical talks around a unifying theme of the “Materials Genome Initiative” (MGI), which is designed to drastically accelerate advanced material design and manufacture via modeling and simulation subsequently coupled with fabrication and characterization of the engineered materials …see attached for more on all that…

MGI is part of the overarching White House OSTP/PCAST “Advanced Manufacturing Partnership” (AMP). Georgia Tech President Bud Peterson is on the AMP Steering Committee.  The first regional AMP mtg. will also be held at Georgia Tech on October 14th.

Both the AMP meeting & MADE@GT are co-located in the GT-Hotel on Oct. 14.  Attendees at AMP can attend MADE@GT, and vice versa.  There will also be a poster session devoted to topics mutual between AMP & MADE@GT.

Information of relevance…..

 

(MADE@GT agenda follows)

—————–Joint Session with AMP——————

Georgia Tech Hotel & Conference Center

800 Spring St. NW; Atlanta, GA 30308

 Oct. 14, 2011

8:00 am  Registration

Continental breakfast

9:00 am  Welcome and Overview

Joseph J. Ensor, Vice President and General Manager for Engineering, Manufacturing & Logistics/Electronic Systems Sector, Northrop Grumman Corp.

G.P. “Bud” Peterson, President, Georgia Institute of Technology (Georgia Tech)

9:45 am  Overview of Workstreams and Breakout Session Instructions

Steve Cross, Executive Vice President for Research, Georgia Tech

Kevin Kolevar, Director of Public Policy and Issues, Dow Chemical Company

10:00 am  Government Panel and Presentations

Moderator, Chuck Thorpe, Assistant Director for Advanced Manufacturing and Robotics, Office of Science and Technology Policy (OSTP)

Department of Defense (DOD); Brett Lambert, Deputy Assistant Secretary of Defense for Manufacturing & Industrial Base Policy

Department of Energy (DOE); Leo Christodoulou, Program Manager, Industrial Technologies Program

National Institute of Standards and Technology (NIST); Michael F. Molnar, Chief Manufacturing Officer

National Science Foundation (NSF); Steven H. McKnight, Director, Civil, Mechanical, and Manufacturing Innovation (CMMI) Division of the Directorate for Engineering

10:45 am  Break

11:00 am  Dr. Cyrus Wadia — (White House OSTP)

“Materials Genome Initiative”

 11:30 am  Dr. Jim Roberto — (ORNL)

“Computational Materials Science for Innovation.”

12:00 pm  COMPLIMENTARY LUNCH + Posters

12:45 pm  Dr. Alex Zunger — (UC-Boulder)…

“Design of materials with target properties”

1:15 pm  Dr. Eui Lee – (NAVAIR/PAX)…

“Strategic planning on future Naval Aviation Materials systems”

1:45 pm  Dr. Scott Sheppard – (Cree)…

“GaN Materials design and implementation for RADAR applications”

2:15 pm  Prof. Ken Sandhage – (GT-MSE)

“Materials Alchemy: Changing the Chemistries, but not Shapes, of 3-D Nanostructured Biogenic and Synthetic Assemblies”

2:45 pm  BREAK

3:00 pm  Prof. David McDowell — (GT-ME)

“Emerging Trends in integrated design of materials and products”

3:30 pm  Prof. David Sholl — (GT-CHBE)…

“Finding needles in a haystack: Selecting high performance nanoporous materials via screening of thousands of structures”

4:00 pm  Prof. Mo Li — (GT-MSE)

“Building Digital Bridges: Linking Microstructures and Material Properties”

4:30 pm  Dr. Jason Nadler – (GTRI-EOSL)

“Enhanced Heat Transfer via Microtemplated Architectures”

5:00 pm  Closing Reception with Drinks/Apps.

 

 

Kindest Regards,

– Jud

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Jud Ready, Ph.D.

Principal Research Engineer & Adj. Professor

Georgia Tech Research Institute

W – 404-407-6036

F – 404-407-9036

http://nano.gtri.gatech.edu/

XSEDE Scholars: includes travel to SC11, applications due Fri Sep 30

XSEDE Scholars Program

— Includes travel grants for the SC11 supercomputing conference in Seattle WA Nov 12 – 16 2011!

Applications due Fri Sep 30 2011

http://bit.ly/xsede_2011

 

An outstanding student opportunity in the computational sciences!

Supercomputers, data collections, new tools, digital services, increased productivity for thousands of scientists around the world.

Sound exciting? Apply to become an XSEDE Scholar today!

XSEDE is a five-year, $121 million project supported by the National Science Foundation that replaces and expands on the NSF’s TeraGrid project.

More than 10,000 scientists used the TeraGrid to complete thousands of research projects, at no cost to the scientists.

XSEDE continues that same sort of work — only in more detail, generating more knowledge, and improving our world in an even broader range of fields.

You can become involved in XSEDE, too, if selected as a Scholar.

XSEDE Scholars will:

  • Attend the SC11 (supercomputing) conference in Seattle, November 12-16, 2011, through travel grants.
  • Meet other XSEDE Scholars in special sessions at SC11.
  • Participate in at least four activities with other Scholars during the year (e.g., technical training, content-based and mentoring webinars).
  • Network with leaders in the XSEDE research community.
  • Learn about research, internships, and career opportunities.

The XSEDE Scholars Program is directed by Richard Tapia and managed by Alice Fisher, Rice University.

 

Apply now at:

http://bit.ly/xsede_2011

 

 

APPLICATION DEADLINE: Friday, September 30, 2011

Underrepresented minority undergraduate or graduate students interested in the computational sciences are especially

encouraged to apply for the year-long program.

 

Questions about the XSEDE Scholars?

Contact Alice Fisher: afisher@rice.edu

 

The Extreme Science and Engineering Discovery Environment (XSEDE) is supported by the National Science Foundation.

XSEDE is led by the University of Illinois’ National Center for Supercomputing Applications. The partnership includes: Carnegie Mellon University/University of Pittsburgh, Cornell University, Indiana University, JüSupercomputing Centre, National Center for Atmospheric Research, The Ohio State University, Purdue University, Rice University, Shodor Education Foundation, Southeastern Universities Research Association, University of California Berkeley, University of California San Diego, University of Chicago, University of Illinois at Urbana-Champaign, University of Tennessee Knoxville, University of Texas at Austin, and the University of Virginia.

 

Find out more about XSEDE at:

https://www.xsede.org/