IT Maintenance 12-27-12
Overview
As part of electrical upgrades to support the next generation of high speed networking, there will be short disruptions to LBL's internet and local network connectivity on Dec 27th from 6:30AM to 1:30pm. During these outages, all services which rely on LBL networks will be unavailable including LBL websites, email, collaborative tools, business applications, and remote access. The outages are anticipated at the beginning and end of the scheduled work.
The outage will not cause a loss of email, which will be queued at the sending location and delivered, albeit delayed, when our email systems systems come back online.
Services Impacted
The maintenance directly impacts the following services:
- All Connectivity between LBL and the internet
- Some internal subnets (see more below)
- All services which rely on network connectivity, including Google Apps (gmail/calendar/etc), network fileshares, cluster computing, remote access, etc.
Why is Gmail impacted?
There are two ways Gmail is impacted during this outage. The first is the indirect impact due to Gmail's dependency on LBL LDAP for authentication. Since Google's servers will not be able to communicate with LBL's servers, users will be unable to authenticate to Gmail. If you are already connected to Gmail during the outage, you will not be disconnected. However, you will not receive new email nor will new emails you create be delivered due to the second impact. Although Lab email is delivered to Gmail and you check it at Gmail, the email is routed through (passes through) Lab email routing systems, for additional security filtering and list processing before being sent to Gmail. In summary, you may be able to remain logged in and read your gmail during the outage, but new mail you receive and send will not be delivered to you until the outage is over.
What will improve after the outage?
This electrical upgrade is necessary to to support the eventual upgrade of our connection to the internet (via ESnet) to 100G - enabling future high speed science data flows and enhanced worldwide collaboration. The actual upgrade will occur in early 2013, this outage is only to install the electrical connections necessary for the upgrade.
How can I get more information?
If you have any questions, feedback, or just want more information, please contact the IT Help Desk at http://help.lbl.gov, or 510-486-4357.
Where can I get the gritty details?
On the IT Outage Page
Your Feedback
Berkeley Lab's HPC Services consultant Yong Qin won the FX10 Championship hosted by Kyushu University at SC12 last month.
The FX10 championship is a competition for performance efficiency on your own code on 12 compute nodes of Fujitsu PRIMEHPC FX10, a commercial version of the K-computer (#3 of the TOP500 November 2012 list) equipped with the SPARC64(TM) IXfx processor and Tofu interconnect. Contestants submitted their codes to the Kyushu University staff and it was subsequently compiled and profiled to measure efficiency. The person with the highest efficiency wins.
According to Professor Keiichiro FUKAZAWA of Kyushu University, any code with an efficiency better than 10% is good. The application that Yong brought in was a code highly optimized for undulator radiation spectrum calculation that we collaborate with the Advanced Light Source (US) and Hiroshima Synchrotron Radiation Center (Japan). The code achieved an astonishing 53% efficiency. The 2nd place winner was only able to reach a 20% efficiency.
Yong attributes his ability to win based on his efforts to greatly reduce the memory footprint and to optimize the code with advanced parallelization techniques. It also helped that he developed this code to run on the newly available 37TF 108-node Lawrencium LR3 cluster which is equipped with 16 Intel Sandybridge processor cores per node - the same number of cores as on the FX10 nodes.
