CablingNetworkNode

Articles

Talon2 online

Hi HPC Community,

Talon2 is back online and compute services have been resumed. All jobs that were suspended appear to have resumed correctly. As always, please notify us immediately if anything looks out of place.

Please note that we are still solving the Lustre Metadata high-availability issue. Lustre is extremely robust but this issue could cause loss of redundancy. However, we feel the risk of this is low enough that we can safely resume  service. 

A few more updates:

Talon2 Maintenance status

Hi,

In regards to the scheduled maintenance.. Unfortunately we encountered a few unforeseen issues that set us back yesterday. However, we are about 80% complete with the work needed in order to restore services. We are fairly confident that Talon2 will be back online this afternoon/evening. Please standby for updates..

To give a scope of the work already accomplished:

- > 200 fiber cables were individually prepared and labeled

- > 100 fiber cables had to be untangled, sorted, and tested

Beginning Maintenance on Talon2

Research Community,
Momentarily we will begin maintenance on the Infiniband infrastructure that you were notified of last Tuesday.  The maintenance is required in order to finalize the upgrade to Talon3.  Therefore, shortly Talon2 (talon2.acs.unt.edu) will temporarily taken offline to perform the maintenance. Compute services will be unavailable until Wednesday afternoon January 25, 2017. 

As always, please message hpc-admin@unt.edu if you have any concerns or questions.

Talon2 Maintenance

HPC Research Community,

Next Monday January 23, 2017, we plan to perform crucial maintenance on the Infiniband infrastructure that supports high speed communication on Talon2.  The maintenance is required in order to finalize the upgrade to Talon3. Unfortunately Talon2 and it's Infiniband network must be temporarily taken offline to perform the maintenance. Compute services will be unavailable from Monday morning on January 23, 2017 until Wednesday afternoon January 25, 2017. 

Talon 3.0 Upgrade Plan

HPC Community,

UIT HPC Services is in the final stages of preparing to upgrade the Talon 2.0 HPC system to Talon 3.0.  This upgrade will result in additional computational capacity while minimizing, as much as possible,  complete interruption of HPC service. However, during the next two months of November and December, expect operations to continue with degraded compute capacity. This will likely increase queue wait times for submitted jobs.

Storage Upgrade: DDN SFA770X

The UNT HPC team has deployed another high-speed parallel filesystem affectionately named scratch2. It is running Lustre 2.5.2 on a DDN SFA7700X storage appliance. The SFA7700X is a hybrid flash storage appliance, purpose-built for Big Data requirements that enables our researchers to continue solving bigger problems. This filesystem has 1.4PB total available and can sustain speeds of up to 10GB/s over FDR Infiniband. For more info see: http://www.ddn.com/products/lustre-file-system-exascaler/

Vi Tips & Tricks

Scott's VI Tips and Tricks - Be more productive with less key strokes!!!

MODES:
Command  This is where you issue commands like movement, editing, ...
:    issue command

Insert
i,I    insert before character/line
a,A    insert after character/line
o,O    insert after/before line
s,S    delete character/line then insert
C      delete rest of line then insert

Replace
R      replace current text

Visual
v      highlight by character
V      highlight by line

Replacing Talon

"At the February 14 meeting of the UNT Board of Regents, a proposal was approved to enter into a three-year lease with Dell for the replacement of our current HPC equipment.  Coordination between UNT's Research and Economic Development, University Information Technology, and UNT's Finance and Budget office has resulted in a funding model that establishes a regular 3-year cycle to replenish HPC technology in order to keep researchers competitive while providing a base level of broad usage HPC technology as a utility for the entire campus."  - Dr. Philip Baczewski, Senior Director of Academic Computing and User Services and Deputy Chief Information Officer for University Information Technology

Read more of the March 2013 Article "UNT Plans High Performance Computing Upgrade" at the UNT Benchmarks Online.