[SC2] HPC resources available to U-M students

By | Educational, Flux, SC2 | No Comments

Brock Palen, Associate Director of Advanced Research Computing-Technology Services, joined the SC2 to talk about all the high performance computing (HPC) resources available to U-M graduate and undergraduate students. A summary of his presentation is here.

Resources:

Available at/through Michigan

  1. Flux for Undergraduates: Undergraduates can use the local flux computing cluster FOR FREE! Please visit the page for more information
    • ARC-Connect: use for Jupyter notebooks and VNC (remote desktop) access of flux resources, useful for remote visualization of big data or just getting a feel for working on linux and flux.
  2. Amazon Web Services: Michigan students get $100/year in amazon web services. While not as cost-effective for some things, very good resource to be aware of.
  3. Hadoop: Michigan’s Hadoop cluster is available for free (I believe you have to apply/demonstrate a need, but you don’t have to pay if it’s accepted). This upcoming workshop will go over the basics, read more if you are interested.

Available via Grant

Brock has an up-to-date webpage linking to and detailing various resources you can apply for.

Highlights:
  1. XSEDE:
    • Startup and teaching allocations are available continuously
    • Research allocations accepts 4x/year
  2. Great Lakes Consortium:
    • Alternate way to get some time on Blue Waters
  3. Amazon/Microsoft/Google:

SC2banner1 copy 2

Research highlights: Running climate models in the cloud

By | General Interest, News, Research | No Comments

Xianglei Huang

Can cloud computing systems help make climate models easier to run? Assistant research scientist Xiuhong Chen and MICDE affiliated faculty Xianglei Huang, from Climate and Space Sciences and Engineering (CLASP), provide some answers to this question in an upcoming issue of Computers & Geoscience (Vol. 98, Jan. 2017, online publication link: http://dx.doi.org/10.1016/j.cageo.2016.09.014).

Teaming up with co-authors Dr. Chaoyi Jiao and Prof. Mark Flanner, also in CLASP, as well as Brock Palen and Todd Raeker from U-M’s Advanced Research Computing – Technology Services (ARC-TS), they compared the reliability and efficiency of Amazon’s Web Service – Elastic Compute 2 (AWS EC2) with U-M’s Flux high performance computing (HPC) cluster in running the Community Earth System Model (CESM), a flagship climate model in the U.S. developed by the National Center for Atmospheric Research.

The team was able to run the CESM in parallel on an AWS EC2 virtual cluster with minimal packaging and code compiling effort, finding that the AWS EC2 can render a parallelization efficiency comparable to Flux, the U-M HPC cluster, when using up to 64 cores. When using more than 64 cores, the communication time between virtual EC2 nodes exceeded the communication time in Flux.

Until now, climate and earth systems simulations had relied on numerical model suites that run on thousands of dedicated HPC cores for hours, days or weeks, depending on the size and scale of each model. Although these HPC resources have the advantage of being supported and maintained by trained IT support staff, making them easier to use them, they are expensive and not readily available to every investigator that needs them.

Furthermore, the systems within reach are sometimes not large enough to run simulations at the desired scales. Commercial cloud systems, on the other hand, are cheaper and accessible to everyone, and have grown significantly in the last few years. One potential drawback of cloud systems is that the user needs to provide and install all the software and the IT expertise needed to run the simulations’ packages.

Chen and Huang’s work represents an important firstxiangleihuangpost2016 step in the use of cloud computing in large-scale climate simulations. Now, cloud computing systems can be considered a viable alternate option to traditional HPC clusters for computational research, potentially allowing researchers to leverage the computational power offered by a cloud environment.

This study was sponsored by the Amazon Climate Initiative through a grant awarded to Prof. Huang. The local simulation in U-M was made possible by a DoE grant awarded to Prof. Huang.

Top image: http://www.cesm.ucar.edu/

2015-2016 Education Snapshot

By | Educational | No Comments

XSEDESummerBootCamp2016

Students at the U-M satellite site of the XSEDE 2016 Summer Bootcamp

We have over 80 students between our Ph.D. in Scientific Computing and the Graduate Certificate in Computational Discovery and Engineering. The students come from five different schools and colleges, and 30 percent are women. We also have partnered with the Multidisciplinary Design Program to offer our Masters students the experience to work on industrial projects and gain practicum credits.

Our faculty have designed two courses that are being offered for the first time: Methods and Practices of Scientific Computing in Fall 2016 and Data-Driven Analysis and Modeling of Complex Systems in Winter 2017. Methods and Practices of Scientific Computing has gathered a tremendous amount of interest, and very quickly was over-subscribed. Data-Driven Analysis and Modeling of Complex Systems is a fast paced research area that combines scientific computing with big data to improve the existing models’ accuracy and representation of physical and biological systems.

Scientific Computing Student Club social gathering

Scientific Computing Student Club social gathering

We have brought a large community of students together by sponsoring and helping found the Scientific Computing Student Club. Its goal is to become a place for all students that use or want to use high performance computing to meet, share ideas, and find peer-to-peer help. It started in February, with social gatherings, talks from expert speakers, and more. The club has nearly 200 members, including undergraduates, graduate students, and postdocs from six U-M schools and colleges.

2015-2016 Education Snapshot

By | Educational | No Comments

XSEDESummerBootCamp2016

Students at the U-M satellite site of the XSEDE 2016 Summer Bootcamp

We have over 80 students between our Ph.D. in Scientific Computing and the Graduate Certificate in Computational Discovery and Engineering. The students come from five different schools and colleges, and 30 percent are women. We also have partnered with the Multidisciplinary Design Program to offer our Masters students the experience to work on industrial projects and gain practicum credits.

Our faculty have designed two courses that are being offered for the first time: Methods and Practices of Scientific Computing in Fall 2016 and Data-Driven Analysis and Modeling of Complex Systems in Winter 2017. Methods and Practices of Scientific Computing has gathered a tremendous amount of interest, and very quickly was over-subscribed. Data-Driven Analysis and Modeling of Complex Systems is a fast paced research area that combines scientific computing with big data to improve the existing models’ accuracy and representation of physical and biological systems.

Scientific Computing Student Club social gathering

Scientific Computing Student Club social gathering

We have brought a large community of students together by sponsoring and helping found the Scientific Computing Student Club. Its goal is to become a place for all students that use or want to use high performance computing to meet, share ideas, and find peer-to-peer help. It started in February, with social gatherings, talks from expert speakers, and more. The club has nearly 200 members, including undergraduates, graduate students, and postdocs from six U-M schools and colleges.

2015-2016 Outreach and Industrial Engagement Snapshot

By | General Interest | No Comments

High school students at miRcore’s GIDAS 2016 Biotechnology Summer Camp working on Flux

We have been working closely with miRcorea nonprofit that supports genetics research. MICDE sponsors compute cycles on Flux for high school students learning how to conduct computational biology research. We have also approached the Society of Women Engineers, U-M chapter, with the goal of  promoting scientific computing through their outreach programs.  

Intel, Cray and Altera experts talking about FPGAs and Scientific Computing

 

 

Our connections to industry have increased and we are constantly exploring new opportunities. Our ConFlux cluster architecture was designed in collaboration with IBM’s experts, and our work with them still continues. In May, MICDE hosted an internal workshop on Field Programmable Gate Arrays (FPGAs) in collaboration with experts from Intel and Cray. We have established a software partnership with MSC Software to enable access for MICDE student projects. Together with the U-M Business Engagement Center, we are creating an affiliates program for companies to invest in our research.

 

 

2015-2016 MICDE Research Snapshot

By | Research | No Comments

karthik1reduced

Professor Karthik Duraisamy (U-M Aerospace Engineering) demonstrates data-driven turbulence modeling.

The Center for Data-Driven Computational Physics was established as a place to concentrate data-driven modeling research across campus. Its activities are focused on ConFlux, a $3.5M groundbreaking cluster funded by NSF with a unique architecture that connects big-data with traditional HPC clusters. ConFlux went online in April, and several teams are already using it, with five projects participating, totaling more than $3M to advance data-driven modeling. Soon, we expect to announce even more successes that are directly attributable to our pioneering role in this research area. 

The Center for Network and Storage-Enabled Collaborative Computational Science was established to tackle the challenges of extracting scientific results collaboratively from large, distributed or diverse data. This research center is a product of the Open Storage Research Infrastructure (OSiRIS), a $5M multi-institutional NSF investment, and is led by MICDE affiliated  research faculty Shawn McKee.

We hosted 16 internationally known speakers in our seminar series, and had a very successful symposium. With speakers including NSF’s ACI Director Irene Qualters, Tom Hughes from ICES, James Sethian from UC Berkeley, Charbel Farhat from Stanford, and Peter Haas from IBM, these events outlined top priorities in our fields, latest research and computing infrastructure, and increased  awareness of the quality and trend-imposing nature of research activities going on at U-M.

2016-02-05jimbelak

Jim Belak (Lawrence Livermore National Laboratory) delivers a talk titled “Preparing for the Future of Computing: Bridging Scales within the Exascale Materials Co-design Center” as part of MICDE Winter 2016 Seminar Series.

MICDE is coordinating or supporting several large proposal submissions to federal agencies. We offer institutional support and our established educational programs to the faculty teams writing these grants. With the backing of our parent unit, Advanced Research Computing, and their technical and consulting services (ARC-Technology Services, and Consulting for Statistics, Computing and Analytics Research), our proposals have proven stronger by virtue of this support in place behind them.

MICDE also is working with the academic units at U-M to identify compelling new directions for hiring faculty who will drive computational science in the future, and supporting these hiring processes. Many of these blue-sky ideas have come from thematic, faculty-led workshops, which we will continue to organize.

NSF EAGER award to study new information and communication technologies in shared connected vehicles

By | News, Research | No Comments

social_networkMICDE associate director Siqian Shen (PI) will collaborate with Co-PIs Tawanna Dillahunt and Tanya Rosenblat from U-M School of Information to conduct interdisciplinary research for a newly announced NSF EArly-concept Grant for Exploratory Research (EAGER) project.

The goal is to investigate the feasibility, challenges, and opportunities of deploying shared connected vehicles with new information and communication technologies (ICTs), to deliver goods and services in future smart & connected communities (S&CC). In taking on a living-lab approach, the study will engage industry, non-profit partners, and underserved populations in Detroit throughout each phase of the project.

The end result will be 1) improved mathematical models and efficient algorithms for optimizing resource allocation, supply-demand matching, and barrier-free vehicle & ICT operations in centralized and decentralized vehicle-and-service-sharing (V&SS) systems; 2) an articulation of the types of critical services that have the highest impact and are needed most among underserved communities (e.g., access to better healthcare or jobs).

miRcore’s high school biotechnology camp a success

By | Educational, General Interest, News | No Comments

GIDAS Biotechnology camp’s high school students learning about microRNA targeted predictions using Flux, with support from MICDE and U-M’s Scientific Computing Student Club members.

 

From Aug. 8-12, 2016, MICDE and ARC-TS donated a Flux allocation and computational support to miRcore and its GIDAS’ Biotechnology Camp for high school students. All the students were able to log in the cluster, and use the command line to run RNAhybrid, a tool for finding the minimum free energy hybridization of a long and a short RNA. The students learned about microRNA target predictions that complemented the camp’s wet lab experiments. Scientific Computing Student Club members Joe Paki and Blair Winograd provided support to the students.

MICDE is partnered with miRcore, a non-profit organization whose mission is to democratize medical research by building funds for microgrants to support innovative genetic research.

MICDE parnters with non-profit miRcore

By | Uncategorized | No Comments

MICDE has partnered with miRcore, a non-profit organization whose mission is to democratize medical research by building funds for microgrants to support innovative genetic research.

MICDE is providing computational resources and support for miRcore outreach activities, as well as connecting our faculty to the miRcore team to provide expertise and to teach students about personalized medicine and end-user driven research.

IMG_20160604_150320

MICDE affiliated Prof. Barry Grant at GIDAS 2016 Research Conference poster session

On June 6, MICDE affiliated faculty member Barry Grant joined miRcore’s high school club GIDAS (Genes In Diseases And Symptoms) in their 2016 Research Conference. The students had the opportunity to experience a real conference setting that helped build their interest in science. The students had a chance to give a talk about a research project, present a poster and hands-on workshops. The conference was a huge success. You can read the proceedings on miRcore’s site.

From August 8-12, 2016, MICDE and ARC-TS will donate a Flux allocation and computational support to the GIDAS’ Biotechnology Camp for high school students. The donations will provide students the opportunity to become familiar with the Unix command line and get hands-on experience on computational genomics.

New MICDE associate directors named

By | General Interest, News | No Comments

MICDE is pleased to announce the appointment of three new associate directors starting July 1, 2016: Karthik Duraisamy (Aerospace Engineering),  Annette Ostling (Ecology and Environmental Biology) and Siqian Shen (Industrial and Operations Engineering). The associate directors, together with MICDE Director Krishna Garikipati and Assistant Director Mariana Carrasco-Teja, will work on initiatives to help MICDE achieve its goals. These include growing MICDE’s educational programs, the Ph.D in Scientific Computing and the Graduate Certificate in Computational Discovery & Engineering; assembling interdisciplinary research teams and helping them secure funding for their projects; engaging with industrial partners; and coordinating outreach activities.