Blog from January, 2013

We have scheduled a regular maintenance for all the clusters in our supercluster infrastructure on Tuesday, Feb 12th from 9:00 am to 5:00 pm. We are in need of taking this downtime to address some stability issues with our job scheduler which has impacted our users.

User logins and access to the data on the cluster filesystems will be blocked. Job scheduler reservations will be put in place such that there will not be any jobs running in the cluster queues for the duration of the downtime. If you are submitting jobs to your clusters before the downtime make sure you request proper wallclock time such that your job will finish before 9:00 am on Feb 12th or else your job will stay queued until after the downtime and then starts to run. User access to the data transfer node will also be turned off.

All the clusters in our supercluster including Alsacc, ARES, Baldur, Cumulus, Explorer, Hbar, JBEI, JCAP, Lawrencium (Lr1, Lr2 & Lr3 nodes) with Co2seq, Matgen, Ganita, Nanotheory, Esd1, Esd2 condos, MHG, Musigny, Nano, Natgas, Voltaire, Vulcan & Yquem all are affected by this downtime. We request all our users to save their work and close their login shell before 9:00 am on Feb 12th.

We apologize for any inconvenience. After the downtime all cluster services will be restored as before. Email us at [email protected] if you have any questions or concerns.

As part of our ongoing quest to find the right suite of realtime collaboration tools for Berkeley Lab researchers, IT is testing Fuze Meeting with the Biosciences Directorate.  The test will actually be one of if not the largest realtime, interactive, video meetings ever attempted at LBNL (that we know of).

 

Participants need to download the fuze software in advance at https://www.fuzebox.com/products/download  and install it on their computer or tablet.

 

For additional attendee information, consult your email.

 

Software Carpentry Boot Camp

We are offering our fourth installment of the 2 day training session dedicated to teaching scientists how to be better computer users on May 9-10, 2013, 9:00AM-4:30PM, 54-130 (Pers Hall).   The class will cover everything from shell to beginning scientific programming.

To register for the boot camp go to: 

http://go.lbl.gov/sc-bootcamp-may9-10

For detailed information about the training sessions go to:

 http://software-carpentry.org/bootcamps/2013-05-lbl.html

There is no charge for these courses, which are supported as part of IT's commitment to help researchers compute and collaborate more effectively.  

 

Excel Classes

Excel for Scientists will be available again in May, and April will have courses in Advanced Functions and Using "What If"Analysis Tools and Macros.  Check the training page for additional details.

 

 

www outage Wed January 23

There will be a 30 minute outage of www.lbl.gov between 6pm and 8pm on Wed January 23.   A small number of other services are also impacted and customers of those services have been notified.  Other websites, network connectivity, email, and all other services are not impacted by this outage.

 

We are glad to announce the availability of third generation (Lr3nodes in the Lawrencium cluster which is the Lab's institutional scientific computational system available for LBNL PI use. We have recently added 108 new compute nodes each equipped with dual-socket eight-core Intel SandyBridge 2.6Ghz processors (16 cores/node) and 64GB of 1600Mhz memory. They  are connected with the latest high performance, low latency 56gb/s FDR infiniband interconnect, compared to the QDR 40gb/s & DDR 20gb/s interconnects in the earlier generation nodes, and are connected into the same user environment and storage as the Lr1 and Lr2 clusters.

Comparison of available Lawrencium Nodes
Third generation (lr3) nodes - 16 core, 64 GB, FDR 56gb/s infiniband
Second generation (lr2nodes - 12-core, 24GB, QDR 40gb/s infiniband
First generation (lr1) nodes - 8-core, 16GB, DDR 20gb/s infiniband

Any Lawrencium user can now access these new nodes by submitting jobs to the same routing queue as always ("lr_batch") but by specifying the  type of nodes on which you want to run in the "-l" line of your PBS job scripts. For example:

1) To run on the new, third generation (lr3nodes, please specify
#PBS -q lr_batch
#PBS -l nodes=X:ppn=Y:lr3

If you do not specify a type of node (either :lr1 or :lr2 or : lr3), the job will default to using the lr1 nodes.

Also we have a 60 second default walltime configured which means any job submitted to Lawrencium queues without the required walltime will run only for 60 seconds so please make sure you specify walltime in all your jobs.

We hope our users will make good use of these new enhanced resources and get more research done quickly.
Interested and new users can visit the HPC Services web site to learn more and request an account. Please email us at [email protected] if you have any questions. Enjoy.


IT is pleased to present a one day course on Agile Project Management for the Lab Community.   Information from the instructor is below.  

Please register for the course here.  There is no charge to attend.


For just any project these days, uncertainty is certain. The customer will change their mind, a ‘must have’ requirement will be discovered, deadlines will move, or an unexpected technological issue will need to be solved. The agile approach to defining and executing a project allows a team to easily adjust to these changing conditions in order to produce the best work product in the best time frame possible.

This workshop compares the agile iterative project approach to a more traditional gated or “waterfall” approach.    What are the benefits of using an agile approach, as well as the costs and risks?   We will explore the key roles, responsibilities, interactions, and processes that make a successful agile project happen.

Attendees will take part in discussions and demonstrations covering:

      • The what and the why of Agile  the myths and truths of Agile
      • Lean, Kanban, and Scrum  the most common agile frameworks
      • Scrum in a Nutshell – the key roles, activities and work products on a scrum team the most common form of agile development
      • Working with agile requirements, specifications, and documentation
      • Estimation  how much work do we have to do?
      • Creating and applying an agile 'Definition of Done'
      • Communication on an agile team  (even when everyones not in the same room)
      • The power of team self-organization and continuous improvement

By the end of the workshop, attendees will have experienced a broad introduction to the world of Agile – its benefits, tradeoffs, costs, and risks.  Concrete ideas will be shared that attendees can immediately apply individually or on team projects. 

 

Instructor Bio

Chris Sims is a Certified Scrum Trainer (CST), agile coach, and recovering C++ developer who helps software development teams improve their productivity and happiness. Chris is the founder of Agile Learning Labs, as well as the Bay Area Agile Managers’ Support group. He is co-author of The Elements of Scrum and has published over 50 articles on agile topics at InfoQ. Even more of his writing can be found on Agile Learning Labs’ blog. Before starting Agile Learning Labs, Chris made a living in roles such as ScrumMaster, Product Owner, Engineering Manager, Project Manager, Software Engineer, Musician, and Auto Mechanic.

 

Operations Customers:

As you may know, occupants of building 46 are being relocated due to the slide above McMillan road.  Workstation Support (MPSG) and the IT Helpdesk were both located in Building 46.     We wanted to alert you to possible interruptions to service as a result of this relocation.

First, the IT Helpdesk has completed its move to a new facility.  We do not anticipate any interruption of service for the helpdesk.

For workstation support, which is a larger and more complex move, we expect that responses to trouble tickets will be impacted over the next two weeks.   We will make every effort to respond to high priority tickets, but please be judicious in marking a ticket “high priority.”  Normal priority tickets may be delayed until the relocation is complete (approximately 2 weeks).   In addition, computer installs and other maintenance will be rescheduled until after the move.  If you have a scheduled install you will receive separate notification about rescheduling.

Thank you for your patience during this process.   If you have any questions or concerns, please contact us at [email protected]


Additional information about the 46 relocation and the hillside is available here:  http://today.lbl.gov/14973/