Research Computing Newsletter - September 2016
As part of an NSF-funded initiative (ACI 1541170), we are happy to announce that Cindy Martin has joined our team as a Cyberinfrastructure Engineer. In this role, Cindy will work directly with research groups to enable optimal use of the university’s research network, with particular focus on supporting researchers with network intensive requirements demanding superior end-to-end performance. By learning more about researcher needs, she will help [U]Tech to extend and enhance the research network to better meet those needs. Cindy has held a variety of network engineering positions, most recently at Penske Logistics, where her primary responsibility was integrating merged companies into the Penske network, as well as designing networks for new remote locations. We welcome her to the team!
We have completed our latest RedCat expansion, bringing total core count to 3532, and estimated to provide 42 TFLOPS of compute capacity. The new 18 batch nodes feature 96GB of RAM and have dual twelve-core processors. SLURM’s feature attribute “-C dodeca96gb" can be utilized within your job script to request use of these new nodes.
Featured Service - Research Networking
Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway. –Andrew Tanenbaum, 1981
That may have been the story in 1981, but in today’s “big data” era, transfers of a terabyte are not uncommon, and we no longer have to rely on UPS or FedEx to share our data. Our campus network was first built over 25 years ago to serve the needs of the entire university. Since that time those needs have changed as have the technologies used. In particular, researchers need to move increasing amounts of data within the campus network and to/from other institutions on a regular basis.
Networks are an essential part of data-intensive science. With the exponential growth of data acquired through experimentation, networks needs to be multi-purposed, performant, and allow for large data transfers between instruments, collaborators and the computational systems used to process the data. To help with data movement, [U]Tech deployed a research focused network known as a Science DMZ that is available in research intensive buildings across campus. The CWRU Science DMZ implementation has been operational since Summer 2014 and was funded through the National Science Foundation CC-NIE Award 1340938 (Bielefeld, PI). Additionally, the award permitted CWRU to increase the bandwidth to four research intensive buildings and provide 10 Gbps connectivity directly into the research lab. The award also increased our Internet2 and OARnet connection to 100 Gbps. Our newest team member Cindy, will be working with many labs on campus to ensure that these resources are utilized to their fullest potential.
When dealing with large datasets, many researchers use USB drives to transfer data to collaborators. If you use USB drives, issues around security, i.e. human subject data restrictions, and the integrity of the data itself, if the drive is your only copy must be addressed. With the recent improvements in our networking infrastructure, the network serves as an alternative to the normal process of copying data to the drive, and then physically moving the data to your colleague. This process can be time intensive and non-optimal when working with terabyte sized datasets. Newer technologies such as Globus can be utilized to move files between your laptop, lab server, research computing center, national supercomputing facility, or any other storage system, using just a browser. Globus is just one example of technologies that [U]Tech can provide to optimize your usage of the network and transferring large data sets. Understanding your needs is crucial in ensuring that you get the most out of the network. In addition, it also helps with making resource investments in technology. Please contact us if you have any questions or need our support.
Research Computing provides a full catalog of services and support options available to the research community. Please go to http://www.case.edu/utech/research-computing for more information.
How to Build a Questionnaire or Survey in REDCap (Research Electronic Data Capture)
When: Friday, September 9, 2016 12:00 p.m.
Where: UH Lakeside, Room 1400
Come learn how to build a questionnaire and/or survey in UH! REDCap (Research Electronic Data Capture). This database software is a secure web application for building and managing online surveys/questionnaires and databases. REDCap provides audit trails for tracking data manipulation and user activity, as well as automated export procedures for seamless data downloads to Excel, PDF, and common statistical packages (SPSS, SAS, Stata).
UH REDCap will allow the safe, secure storage of all your research data, including the storage of PHI (Protected Health Information) HIPAA identifier fields for your UH IRB protocols.
Presented by Sheree Hemphill, MS DCRU Informatics Manager, CTSC Biomedical Informatics Core - CREC CREDITS: 0
Cleveland Hadoop Users Group - Back to School Mega-Meetup!
When: Monday, September 12, 2016 5:00 p.m.
Where: Cleveland Museum of Natural History
- Dr. Evalyn Gates - Director of Cleveland Museum of Natural History. Dr. Gates wrote the 2009 book "Einstein’s Telescope: The Hunt for Dark Matter and Dark Energy in the Universe."
- IBM Fellow Shiv Vaithyanathan - Machine Learning. Shiv was the creator of Apache SystemML
- Yelp - "Pets to Cattle: Semi-Stateful Services on Ephemeral Infrastructure"
- Cloudera - "Apache Kudu: Fast Analytics On Fast Data"
- Holden Karau - Author of "Learning Spark", and "High Performance Spark". Spark is a framework for big data processing and analysis.
- Case Professor Roger French - Data processing for solar farms
Please RSVP for accurate headcount.
Science Café Cleveland: From Big Data to Big Discoveries -Concert Hall at the Music Box Supper Club
When: Monday, September 12, 2016 5:30 p.m.
Where: Concert Hall at the Music Box Supper Club
William S. Bush, assistant professor of epidemiology and biostatistics, and Aaron Goldenberg, associate professor of bioethics and associate director of the Center for Genetic Research Ethics and Law, will lead the Sept. 12 discussion, titled “From Big Data to Big Discoveries: How Medical Records Can Support Scientific Research.
Please RSVP for accurate headcount.
When: Tuesday, September 13, 2016 9:00 a.m. - 5:00 p.m.
Where: Toepfer Room, Adelbert Hall
The first session in our HPC training series will kick off in September. Topics include introduction to HPC at CWRU, submitting jobs, using MATLAB on RedCat and much more!
Exploring Non-traditional uses of High Performance - Computing Session 1 - What is Big Data?
When: Wednesday, September 28, 2016 12:00 p.m
Where: Clark Hall, Room 206
This introductory session is the first of a four-part series looking at computation in non-traditional areas. The series aspires to explore "Big Data" topics in fields that may not traditionally have used computational methodologies, such as the humanities, social sciences, and law. Each of the four lunchtime meetings during the Fall semester will have two guest speakers -- some showcasing ongoing projects and others who work for Big Data service providers.
Please join us for these informative sessions. Your participation and insight is greatly appreciated. Please RSVP for the first session by September 21 to Lee Zickel at firstname.lastname@example.org.
Seminar in Multidisciplinary Clinical and Translational Research
When: Monday, October 10, 2016 12:00 p.m
Where: Wolstein Research Building, Room 6136
This seminar series will address the processes and challenges of multidisciplinary clinical/translational science, through which discoveries in the laboratory or in early clinical studies are transformed into interventions, treatments, and ultimately, best practices and policies on national and international levels. This seminar will use a case-based approach.
Anant Madabushi, PhD, F. Alex Nason Professor II and Director, Center for Computational Imaging and Personalized Diagnostics, Department of Biomedical Engineering, CWRU, will discuss "Image Computing and Precision Medicine".