Daily Bulletin Archive

Feb. 27, 2017

After several days of intensive work by CISL's Supercomputing Services Group and Consulting Services Group, Cheyenne was released to Accelerated Scientific Discovery (ASD) project users around 9:30 p.m. Wednesday. The teams successfully deployed software to prevent out-of-memory crashes on nodes and a critical update to the PBS Pro scheduler.

CISL will continue to rigorously test the system before it is released to the broader user community. Be sure to watch for updates on Cheyenne's availability in the Daily Bulletin.

Feb. 22, 2017
The major Cheyenne system updates that were started Tuesday morning encountered several unexpected issues late last night, resulting in the maintenance downtime being extended into today.
 
At this time we hope to have Cheyenne back up by midday, when CISL will resume testing before releasing the system for Accelerated Scientific Discovery projects.
 
We will notify the user community on any and all updates by email as soon as more information is available.
 
Feb. 22, 2017

HPSS will require a short two hour downtime this morning to complete some database work related to the upgrade of last week.

Feb. 21, 2017
Major upgrades to the Cheyenne system are scheduled for Tuesday, February 21, and will require an all-day system outage starting at 9 a.m. The most significant upgrades will be to the PBS Pro job scheduler and to the program for detecting out-of-memory (OOM) events on batch nodes.

Cheyenne users should be aware that jobs submitted after 8 p.m. Monday will be held until after the system is rebooted Tuesday night. Because of the significant updates to PBS Pro, there is no guarantee that those jobs will survive the system reboot and may need to be resubmitted.

Feb. 21, 2017

HPSS: Downtime Tuesday, February 21, 9:00 am - 3:00 pm.

Cheyenne: Downtime Tuesday, February 21, Starting at 9:00 am

No Downtime: Yellowstone, Geyser/Caldera

Feb. 17, 2017

By Sophie Hou, NCAR Data Curation & Stewardship Coordinator

The state of being an “unloved data set” is often reached unintentionally over time. It takes a lot of effort and resources to set up a study and to produce the corresponding results. However, if research projects lose sight of data management actions, research results and products could be at the risk of becoming forgotten or “unloved” when the team moves on to new projects.

In order to mitigate the risks of research results from becoming “unloved,” the Digital Asset Services Hub (DASH) and the Data Curation & Stewardship Coordinator offer a wide variety of services, including the DMP Preparation Guidance and Template Document and DMP Checklist for Awarded Proposals.

After all, as the following quote from LYD suggests, we want to help in promoting scientific results and allow them to be used, so that they would be valued for the long term.

“Data that is mobile, visible and well-loved stands a better chance of surviving” ~ Kurt Bollacker

If you would like to learn more about DASH and its services after the LYD week, including how we could help in reviving your unloved data, please contact us at datahelp@ucar.edu. This information is provided as part of the Love Your Data (LYD) Week 2017 series.

Feb. 16, 2017

By Sophie Hou, NCAR Data Curation & Stewardship Coordinator

Finding the right data can depend on many factors, including how the research questions are framed, i.e. the who, what, where, and when that the researchers and their collaborators would like to study.

Additionally, the determination of whether a dataset could be “right” for a research rely on the information that the original data producers and providers were able to document and make available as well as the information that the potential data users are able to access and understand.

Using metadata to document, describe, and define a research product/output, ensuring high quality (i.e. completeness, accuracy, credibility and consistency) of the product/output and its associated metadata, as well as archiving/preserving the product/output at a repository are all actions that could be taken by researchers and their collaborators to facilitate the correct matching of the research questions with the desired, appropriate products/output.

By helping NCAR Labs' results to be more “FAIR” (i.e. Findable, Accessible, Interoperable, and Re-usable) through the development and implementation of a search and discovery system and a general repository, the Data Stewardship Engineering Team is also looking to improve the amount of right data from NCAR that could be used to in a broad range of scientific studies.

To learn more about the upcoming DASH Search and Discovery system and the Repository, please check the Digital Asset Services Hub (DASH) website (updates will be provided over the next months) or contact us at datahelp@ucar.edu. This information is provided as part of the Love Your Data (LYD) Week 2017 series.

Feb. 15, 2017

By Sophie Hou, NCAR Data Curation & Stewardship Coordinator

Similar to data quality, data’s “goodness” can also be challenging to quantify and measure. However, a dataset that demonstrates a higher degree of quality (i.e. completeness, accuracy, credibility and consistency) could also potentially be “better” than its peer datasets that show lesser degree of these attributes.

For LYD, “good data” should also be “FAIR,” i.e. Findable, Accessible, Interoperable, and Re-usable, and in order to be “FAIR,” in addition to using metadata to document, describe, and define a research product/output, such as a dataset, publication, software, and model, it is also crucial to deposit and archive/preserve the product/output, so that it will not be lost and could be taken care of for the long term.

In order to help enable NCAR Labs' results to be “FAIR,” the Data Stewardship Engineering Team is developing and implementing a search and discovery system and a general repository that will be made available through the Digital Asset Services Hub (DASH).

To learn more about the upcoming DASH Search and Discovery system and the Repository, please check the Digital Asset Services Hub (DASH) website (updates will be provided over the next months) or contact us at datahelp@ucar.edu. This information is provided as part of the Love Your Data (LYD) Week 2017 series.

Feb. 15, 2017

Critical updates to the Cheyenne system that were scheduled for today have been postponed because of unexpected problems encountered during tests of the PBS Pro scheduling system on the Laramie cluster. Determining the causes of the problems and implementing fixes may take several more days. Once the fixes are installed and tested on Laramie, CISL will reschedule the Cheyenne updates and make the system available as soon as possible to the general user community.

We will continue to keep you up to date with any and all new information on Cheyenne’s status

Feb. 14, 2017

 By Sophie Hou, NCAR Data Curation & Stewardship Coordinator

 In order to document, describe, and define data, especially with quality attributes (e.g. complete, accurate, credible, consistent), it is helpful for researchers and their collaborators to be familiar with the concept of metadata as well as the commonly known metadata standards and formats associated with the researchers’ fields of study. The use of metadata, or “good documentation,” could tell people that “they can trust your data by enabling validation, replication, and reuse” as LYD suggests.

At NCAR, the Digital Asset Services Hub (DASH) can help researchers in learning about metadata, the different standards and formats, and how to use them. An example is the metadata from DataCite, which is used when requesting and registering digital object identifiers (DOIs). Another example is the metadata format used with the upcoming DASH Search and Discovery system (a centralized system that uses metadata to assist with and enhance the effectiveness/efficiency of finding, understanding, and using research products and output from NCAR labs).

To learn more about DataCite’s metadata and how it works with DOI assignments, please visit DASH’s Digital Object Identifiers (DOIs) page, and for further information about NCAR’s metadata used for DASH Search and Discovery system, please check back at DASH Search and Discovery page (updates will be provided over the next months) or contact us at datahelp@ucar.edu. This information is provided as part of the Love Your Data (LYD) Week 2017 series.

Pages