Explore ARCExplore ARC

The 2018 MICDE Symposium: Summary by Bradley Dice, Ph.D student in Physics and Computational Science

By | Uncategorized

This piece was first published in LinkedIn by Bradley Dice, U-M Ph.D student in Physics and Computational Science.

MICDE Symposium 2018: Computation, A Pillar of Science and a Lens to the Future

High-performance computing (HPC) is becoming an increasingly powerful tool in the hands of scientists, driving new discoveries in physical sciences, life sciences, and social sciences. The development of new (frequently domain-specific) approaches to machine learning and faster, smarter processing of sets of Big Data allows us to explore questions that were previously impossible to study. Yesterday, I presented a poster at the Michigan Institute for Computational Discovery & Engineering (MICDE) annual Symposium and attended a number of talks by researchers working at the intersection of high-performance computing and their domain science. The theme for the symposium was “Computation: A Pillar of Science and a Lens to the Future.”

Collaborative Computational Science with signac

My scientific work, and the work of my colleagues in the Glotzer lab, has been made vastly more efficient through the use of tools for collaborative science, particularly the signac framework. I presented a poster about how the signac framework (composed of open-source Python packages signacsignac-flow, and signac-dashboard) enables scientists to rapidly simulate, model, and analyze data. The name comes from painter Paul Signac, who, along with Georges Seurat, founded the style of pointillism. This neo-impressionist style uses tiny dots of color instead of long brushstrokes, which collectively form a beautiful image when the viewer steps back. This metaphor fits the way that a lot of science works: given only points of data, scientists aim to see the whole picture and tell its story. Since our lab studies materials, our “points” of data fit into a multidimensional parameter space, where quantities like pressure and temperature, or even particles’ shapes, may vary. Using this data, our lab computationally designs novel materials from nanoparticles and studies the physics of complex crystalline structures.

The core signac package, which acts as a database on top of the file system, helps organize and manage scientific data and metadata. Its companion tool signac-flow enables users to quickly define “workflows” that run on supercomputing clusters, determining what operations to perform and submitting the jobs to the cluster for processing. Finally, signac-dashboard (which I develop) provides a web-based data visualization interface that allows users to quickly scan for interesting results and answer scientific questions. These tools include tutorials and documentation, to help users acquaint themselves and get on to doing science as quickly as possible. Importantly, the tools are not specific to materials science. Many scientific fields have similar questions, and the toolkit can easily be applied in fields where exploration or optimization within parameter spaces are common, ranging from fluid mechanics to machine learning.

During the symposium, I learned a lot about how others are using scientific computing in their own work. The symposium speakers came from a wide range of fields, including biology, mathematics, and fluid dynamics. Some of my favorite talks are described below.

The Past: Phylogeny and Uncovering Life’s Origins

High-performance computing is enabling scientists to look in all sorts of directions, including into the past. Stephen Smith, Assistant Professor of Ecology and Evolutionary Biology at the University of Michigan, talked about his lab’s research in detecting evolutionary patterns using genomic data. From the wealth of genetic data that scientists have collected, the Smith lab aims to improve our understanding of the “tree of life”: the overarching phylogenetic tree that can explain the progress of speciation over time. Projects like Open Tree of Life and PHLAWD, an open-source C++ project to process data from the National Center for Biotechnology Information’s GenBank data source, are just two of the ways that open science and big data are informing our understanding of life itself.

The Present: From Algebra to Autonomy

Cleve Moler, the original author of the MATLAB language and chief mathematician, chairman, and cofounder of MathWorks, spoke about his career and how the tools MATLAB has provided for numerical linear algebra (and many other computational tasks) have been important for the development of science and engineering over the last 34 years. MATLAB is taught to STEM students in many undergraduate curricula, and is used widely across industry to simulate and model the behavior of real systems. Features like the Automated System Driving Toolbox are poised to play a role in autonomous vehicles and the difficult computational tasks inherent in their operation.

The Future: Parallel-in-Time Predictions and Meteorology

A significant challenge in weather and climate modeling is that supercomputer architectures are highly parallel, while many simulations of fluids are inherently serial: each timestep must be computed before the next timestep can begin. Beth Wingate, Professor of Mathematics at the University of Exeter and published poet, is developing a powerful approach that may change the way that such models work. Called “parallel-in-time,” it separates the effects of slow dynamics and fast dynamics, enabling parallel architectures to take advantage of longer timesteps and separate the work across many processors.

Conclusions

Computational science is growing rapidly, improving our ability to address the most pressing questions and the mysteries of our world. As new supercomputing resources come online, such as Oak Ridge National Laboratories’ Summit, the promise of exascale computing is coming ever closer to reality. I look forward to what the next year of HPC will bring to our world.

[SC2 Jobs] Gateway Development Internship for SGCI

By | SC2 jobs

The Workforce Development team from SGCI offers eight-week summer internships for students interested in developing their gateway development skills. Participants will be placed at one of the seven universities that form the SGCI partnership, or a specific site can be suggested by an SGCI client, partner, or others who are interested in hosting a student intern.

Eligible applicants include graduate students majoring in computer science or computer engineering (or related fields) at any level and undergraduates majoring in computer science or computer engineering (or related fields) who have completed their junior year and who demonstrate strong programming and software engineering skills.

Participants will receive a stipend, plus housing and transportation. Interns will be required to attend the Gateways 2018 conference, for which Workforce Development will provide funding. Attending PEARC18 is recommended, but not required. Funding will be provided by Workforce Development to interns who decide to attend.

Learn more about the internship opportunity by reading these blog posts written by some of the students who worked as SGCI interns during summer of 2017:

Summer 2018 Gateway Development Internships

If you are a student and would like to apply for an internshipcomplete this application form. The deadline for submitting this application and supporting documents is April 27, 2018. We will begin to fill slots on March 9, 2018, and will continue the review process until all slots are filled.

 

Application deadline

April 27, 2018

[SC2 Jobs] Coding Institute Workshop for SGCI

By | SC2 jobs

The Science Gateways Community Institute will host a Summer 2018 workshop that will be a four-week Coding Institute for undergraduate students that will take place on the campus of Elizabeth City State University. The workshop will cover the core skills needed to be productive in the design and maintenance of science gateways. The program will be presented as short tutorials alternated with practical experiences, and all instruction will be done via live coding.

Eligible applicants are undergraduate students majoring in computer science or computer engineering (or related fields) who have an interest in the design and maintenance of science gateways. Ten participants will be selected for the Coding Institute. Participants of the Coding Institute will receive a weekly stipend of $500 plus funding for transportation and housing. All selected participants will be required to attend both the PEARC18 and Gateways 2018 conferences. Funding to attend both conferences will be provided by SGCI’s Workforce Development.

If you are an undergraduate student who would like to apply for the four-week Coding Institutecomplete this application form. The deadline for submitting this application form and supporting documents is April 27, 2018.

 

Minimum Requirements:
•Undergraduate majoring in Computer Science or Computer Engineering (or related fields) and have an interest in the design and maintenance of science gateways

Job category

Coding Institute Workshop

Location

Elizabeth City, NC

Application deadline

April 27, 2018

ConFlux cluster expands

By | General Interest, Happenings, HPC, News

ARC-TS has installed 15 new compute nodes into the ConFlux cluster. These nodes have the same 20 cores CPU as the original set, but with 256 GB of RAM instead of 128 GB. Neither the original nodes nor the newly added ones contain any GPUs

As a result, jobs should spend less time in queue, and users can be more liberal in their memory requirements.

[SC2 Jobs] SLATE: A Platform for Scientific Cyberinfrastructure

By | SC2 jobs

We are looking for motivated individuals interested in learning cutting-edge technologies to help us develop, prototype and test SLATE: Service Layes At The Edge, a collaborative project between the University of Michigan, University of Chicago and the University of Utah.  The SLATE team is working to deliver a new platform for scientific cyberinfrastructure (see http://slateci.io/). Building upon container-based technologies (e.g. Docker, Kubernetes, Helm, …) we aim to create a distributed environment where scientific collaborations can create, deploy and operate the tools they need to manage their data collection, distribution, and processing.

The successful candidate will help to take existing applications and package them in a way that can be readily used within the platform so that the details of installation, configuration, and operations of the application are taken care of for the scientific user. The candidate will learn how to use the latest container technologies and how to use them to solve concrete problems. The position will pay 12-15$ an hour based on experience for 20-30 hours a week depending on availability of the candidate.

 

Minimum Requirements:

* Basic experience with Linux administration

 

Preferred Requirements:

* Experience with source control (e.g. git)

* Experience with virtual machine technology (e.g. VirtualBox)

* Experience with container technology (e.g. Docker)

 

If interested, please send your CV to Brock Palen brockp@umich.edu

 

HPC training workshops begin Tuesday, Feb. 13

By | Educational, Events, General Interest, Happenings, HPC, News

series of training workshops in high performance computing will be held Feb. 12 through March 6, 2018, presented by CSCAR in conjunction with Advanced Research Computing – Technology Services (ARC-TS).

Introduction to the Linux command Line
This course will familiarize the student with the basics of accessing and interacting with Linux computers using the GNU/Linux operating system’s Bash shell, also known as the “command line.”
Location: East Hall, Room B254, 530 Church St.
Dates: (Please sign up for only one)
• Tuesday, Feb. 13, 1 – 4 p.m. (full descriptionregistration)
• Friday, Feb. 16, 9 a.m. – noon (full description | registration)

Introduction to the Flux cluster and batch computing
This workshop will provide a brief overview of the components of the Flux cluster, including the resource manager and scheduler, and will offer students hands-on experience.
Location: East Hall, Room B254, 530 Church St.
Dates: (Please sign up for only one)
• Monday, Feb. 19, 1 – 4 p.m. (full description | registration)
• Tuesday, March 6, 1 – 4 p.m. (full description | registration)

Advanced batch computing on the Flux cluster
This course will cover advanced areas of cluster computing on the Flux cluster, including common parallel programming models, dependent and array scheduling, and a brief introduction to scientific computing with Python, among other topics.
Location: East Hall, Room B250, 530 Church St.
Dates: (Please sign up for only one)
• Wednesday, Feb. 21, 1 – 5 p.m. (full description | registration)
• Friday, Feb. 23, 1 – 5 p.m. (full description | registration)

Hadoop and Spark workshop
Learn how to process large amounts (up to terabytes) of data using SQL and/or simple programming models available in Python, R, Scala, and Java.
Location: East Hall, Room B250, 530 Church St.
Dates: (Please sign up for only one)
• Thursday, Feb. 22, 1 – 5 p.m. (full description | registration)

2016-2017 MICDE Research Snapshot

By | Research

2016-2017 has been a year of sustained growth for MICDE’s research portfolio. The number of faculty affiliated with the institute stands at 130, spanning 30 departments and eight schools and colleges. The Center for Scientific Software Infrastructure was established to bring together the U-M community engaged in developing open scientific software. It will focus on establishing best practices for developing, disseminating and documenting scientific software in the public domain. Led by Prof. Emanuel Gull (Physics), the Center aims to provide training and support for researchers that are ready to transform their research codes into well-engineered software. It offers grant support in the form of programmers, consultants, and administrative assistance. It includes a portal to share your code with the research community at large.

MICDE’s two established centers, the Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) and the Center for Data-Driven Computational Physics (DaCoP), each held their first symposium, showcasing their first year of research activities. This included evidence of the growing reach of OSiRIS, the open framework for storage, computation and collaboration against big scientific data, and the first results from ConFlux, U-M’s groundbreaking computing cluster for data-driven computational physics. These results have been presented at several conferences, and are appearing in the leading computational journals.

Vorticity field at a late time in the evolution of an elliptic vortex patch computed by a Lagrangian particle method with remeshing and treecode-accelerated evaluation of the Biot-Savart integral. (source: Ling Xu)

MICDE also funded its first round of Catalyst Grants, that are supporting four innovative computational science research projects. Research funded by the Catalyst Grants is breaking new ground, while helping define the future of computational science. This research consists of:

  • studies of the neuronal dynamics of learning and memory formation;
  • new algorithms for the complex, nonlinear dynamics of power grids;
  • novel integral equations methods using recent advances in numerical analysis;  
  • and probabilistic computational frameworks for rare but often catastrophic events.

The past academic year MICDE hosted 14 external speakers with backgrounds and research concentrations that span the breadth of computational science of today and the future. The series culminated in MICDE’s annual symposium: “The New Era of Data-Enabled Computational Science,” which featured talks by worldwide leaders in computational science, including U-M faculty. The symposium included a student poster competition with over 50 entries.

Dr. Ann Almgren from the Lawrence Livermore National Lab speaking about Next Generation AMR, part of the 2016-2017 MICDE Seminar Series

MICDE faculty are committed to growing the already strong U-M community of computational scientists. Over the past year, as before, we have organized a number of workshops to foster collaboration and put together interdisciplinary teams in response to funding calls from federal agencies and foundations.   MICDE offers faculty teams institutional support and direct links to our excellent educational programs as well as cyberinfrastructure, all of which strengthen faculty proposals. With the backing of our parent unit, Advanced Research Computing (ARC), and its technical and consulting services (ARC-Technology Services, and Consulting for Statistics, Computing and Analytics Research), this effort has raised over $22M in external funding over the past 2 years. This includes support from federal agencies (NSF, NIH, and DOD), as well as from industry.  We also work with the academic units at U-M to identify compelling new areas for recruiting the type of faculty members who will drive computational science in the future.

 

2016-2017 Education Snapshot

By | Educational, General Interest, News

Over the past year, MICDE’s educational programs and activities have experienced tremendous growth. The Graduate Certificate in Computational Discovery and Engineering currently has 50 students enrolled, spanning 19 departments from 5 different schools and colleges. Sixteen students graduated within the last academic year, and 44 have graduated since the Graduate Certificate was established in 2013. Even further, the number of women in the program went from zero in 2014 to 15 currently enrolled.

The Ph.D. in Scientific Computing has experienced extraordinary growth, with 74 students enrolled from 20 departments, and four schools or colleges. We added a section to our web site with both programs’ alumni information.

We are working to broaden as well as to deepen the activities and resources available to students in both programs. Twenty MICDE fellowships were awarded this academic year to students in our programs. We continued to sponsor student software teams at competitions, as well as individual students presenting their work at leading conferences. On-campus, MICDE student activities include networking lunches, and the Scientific Computing Student Club (SC2). On the programmatic front, our non-engineering students now have access to a CAEN account that gives them more options to connect and use U-M High Performance Computing resources. Relevant grant opportunities for students are tracked and updated in MICDE’s grant webpage

2016-2017 MICDE Fellow Yuxi Chen (ClaSp) presenting his work at the MICDE Annual Symposium

Several educational projects and initiatives are afoot at MICDE, including a Massively Open Online Class (MOOC) in Computational Thinking targeting both high school students and their teachers. This MOOC aims to introduce learners to algorithmic approaches to problems. This initiative is being developed in collaboration with the School of Education, the office of Academic Innovation, and with input from a number of high schools in the Detroit Metropolitan Area.The two new courses launched by MICDE faculty last year, Methods and Practices of Scientific Computing, and Data-Driven Analysis and Modeling of Complex Systems, were successful in their first offerings during the 2016-2017 academic year, and are being offered again in 2017-2018. Other teams of MICDE faculty are at work across campus to develop new courses in computational science.

2016-2017 Outreach and Industrial Engagement Snapshot

By | General Interest

2017 miRcore’s GIDAS Biotechnology Summer Camp participants

Community Outreach

MICDE remains committed to advancing the understanding of science in general, and computational science in particular, in the community. To this end we have continued our support of internal and external organizations. Externally, our ongoing support of the non profit science outreach group, miRcore, included running MICDE sponsored compute cycles on Flux for high school students participating in miRcore’s computational biology summer camps through their student network called GIDAS. We also continued to support the undergraduate Biosoftware Team that has competed in the International Genetically Engineered Machine (iGEM) year competition for the past five years. The team participates in the software track aimed for computer scientists and developers to nurture their knowledge of biology, and for computational biologists, bioinformaticians and biologists to enhance their aptitude for building software. Over the past couple of years, the team has been developing ProtoCat, a software developed to address the issue of reproducibility in synthetic biology. It is a collaborative platform on which researchers share their experiment protocols and users can customize them to meet their own needs. For the third year in a row, the team returned with a gold medal.

2017 BioSoftware Team

Internally, less than two years since its inception, the Scientific Computing Student Club (SC2) has established several activities that complement the formal training in computational science available at U-M, including through MICDE’s PhD in Scientific Computing, and Graduate Certificate in Computational Discovery and Engineering. Over the past year, the SC2 had his own invited speakers, organized tours to the Flux facility and the U-M 3D Lab, organized the first Visualization Challenge, co-sponsored by NVIDIA, and just recently added a section on its web page for academic and non-academic job opportunities. During the 2017 Fall Term, SC2 students ran a weekly Machine Learning Collaborative Workshop, and the group is planning a hands-on series on code parallelization.

Industrial Engagement

We continue working towards increasing our engagement with industry. Over the last two years, in addition to NVIDIA, MICDE has established partnerships with IBM, through the joint design and development of our computer cluster, ConFlux, and with Toyota Research Institute, through a funded project on scientific software for materials research. We are now working in partnership with the U-M Business Engagement Center to create an MICDE industrial affiliates program, which will provide many additional avenues for interaction between our students or faculty and industry.

 

 

 

[SC2 Jobs] Machine Learning Scientist for Toyota Research Institute of North America

By | SC2 jobs

Toyota Research Institute of North America, located in Ann Arbor, Michigan, is seeking a machine learning scientist to support the in-house research activities. This individual will join a team responsible to develop state-of-art methodologies for material informatics. The position requires staying abreast of emerging field of machine learning, performing original research, publishing/presenting results, involve in collaborative research. Candidate must be able to work effectively with a diverse group of scientists.
The position is subject to annually contract renewal.

Key Responsibilities:
• Develop machine learning models to deal with problems/challenges in material informatics;
• Establish tools to collect and structure materials data and harvest valuable information
subsequently;
• Perform text mining from scientific literatures and internal technical documents;
• Frequently communicate with materials scientist within the organization;
• Effectively respond to the challenges emerging in materials project;
• Regularly report and present to the research team and managements;

Minimum Requirements:
• MS or above degree in Computer Science, Statistics or related technical field or equivalent
practical experience;
• Strength with machine learning and text mining techniques;
• Fluency in programming languages (Python, C/C++, Java);
• Hands-on experience with statistical software (R, SAS, Matlab, Python);
• Strong verbal and written communication skills;
• Self-motivated, intelligent individual with initiative and drive for overcoming technical
challenges;

Preferred Requirements:
• Experience with deep learning techniques;
• Experience in projects related to materials science, chemistry and physics;
• Established capability in scientific writing and presentation;

 

The applicant should send the resume to chen.ling@toyota.com before February 9, 2018.

Job category

Machine Learning Scientist

Location

Ann Arbor, MI

Application deadline

February 9, 2018