News in Brief
DXC EXPANDS SCOPE
PSC’s Data Exacell (DXC) pilot project has increased the community of researchers who can use the system and the scope of projects it can serve by adding new hardware and software resources. The DXC team added data storage to the system and launched dedicated servers to run Hadoop, the popular data-intensive software environment used in many scientific and business applications.
The DXC team has tested the new storage hardware to identify and eliminate bottlenecks at various stages along the data pipeline. The hardware runs PSC’s SLASH2 file system, which helps remote users manage and analyze large datasets with speeds higher than those on the users’ own hard drives.
In August, PSC staff used the new Hadoop capability to teach a national workshop on Big Data analysis. Hadoop is one of a number of experiments DXC has undertaken to allow researchers to bring previously-successful software solutions to HPC systems with minimal re-tooling and maximal reproducibility.
DXC’s mission is to develop and test new hardware and software technologies, as well as system architectures, to support data-intensive research. These advances will be used at PSC and will be made available as Data Infrastructure Building
Blocks to benefit the National Science Foundation’s user community.
2015 MARC INTERNS COMPLETE SUMMER PROJECTS
Seven summer interns from seven institutions of higher learning have completed bioinformatics projects funded by the fifth year of PSC’s National Institutes of Health (NIH) Minority Access to Research Careers (MARC) program grant.
A collaboration with minority-serving partner institutions, MARC is a special technology transfer and outreach program, begun at PSC in 2001, to increase minority participation in biomedical research. Most recently supported by a 2011, five-year NIH grant, MARC includes:
• An intense two-week Summer Institute in bioinformatics at PSC, open to faculty and graduate students at minority-serving institutions. The session focuses on preparing faculty to teach a semester-long bioinformatics course with a focus on cutting-edge technologies like Next-Generation Sequencing.
• An eight-week research internship at PSC for students that have completed bioinformatics training on their local campus.
• The development of a model curriculum for Bioinformatics including related course materials for the fields of Biology, Computational Science and Mathematics. Elements from these courses are incorporated into the workshop as they are completed.
• Assistance in establishing and strengthening the Bioinformatics programs at two minority-serving campuses each year, including teaching assistance for newly established courses.
MARC intern projects this year included computer investigations of how human populations in different parts of the world vary genetically, with an eye toward identifying disease vulnerability; how genes in commercial chicken breeds affect development of fatty tissues; and generating a knowledge base for retroviruses—a family of viruses distantly related to HIV—in birds.
2015 PSC MARC Interns
Antonio Camacho Flores – Universidad Metropolitana, San Juan, Puerto Rico
Nicholas Torvell Cook – Tennessee State University, Nashville
Jimmaline Hardy – Morehouse School of Medicine, Atlanta
Okoye M. Echezona – University of Texas, El Paso
Ivan L. Jiminez Ruiz – University of Puerto Rico, Piedras Campus
Selena M. Rodriguez Riviera – University of Puerto Rico, Medical Science Campus
Samantha Octavia Whitaker – North Carolina A&T State University, Greensboro
PSC BUILDING EARLY WARNING TOOL FOR FIXING INTERNET TRAFFIC JAMS
A $300,000 NSF grant will enable software engineers at PSC to build the first practical tool for warning individual users and their network administrators when their connection has developed a problem that will slow or halt data flow. The one-year project will build a tool called XSight, intended to become a standard part of the toolbox for maintaining network connections for all operating systems. XSight will build on PSC’s Web10G, a set of software tools for obtaining connection data.
“XSight will represent a proactive approach toward resolving network problems,” says PSC’s Chris Rapier, principal investigator in the project. “It’s different from other approaches to measure network performance because it takes advantage of Web10G’s ability to collect data on individual data transfers, both in the network and in an application’s interaction with the network.”
Previous attempts to measure network performance, while useful for administrators trying to recognize systemic problems, were unable to provide information either timely or detailed enough to help individual users with networking problems. By making use of the data available from Web10G, XSight will overcome these limitations, providing an automated warning system that detects data slowdowns and their causes before they become acute.
XSight will exploit Web10G’s ability to extract data about network connections from TCP/IP, the set of protocols for transferring data that underlie the entire Internet. PSC developed Web10G in collaboration with the National Center for Supercomputing Applications.
The XSight project will be funded through the NSF’s Early-concept Grants for Exploratory Research (EAGER) program. EAGER supports “high risk/high payoff” early stage exploratory work on “untested but potentially transformative research ideas or approaches.”
CHANGING VACCINE VIAL SIZE CAN IMPROVE SUPPLY: PSC/HOPKINS STUDY
Changing vaccine vial size – the total number of doses a single vaccine vial contains – can improve immunization rates and decrease costs, according to a paper recently published in the journal Vaccine.
Researchers at PSC and the International Vaccine Access Center at the Johns Hopkins Bloomberg School of Public Health used their supply-chain simulation tool HERMES to estimate the impact of various vaccine vial sizes on the vaccine supply chain, including vaccine availability and costs. HERMES is a software platform created by the HERMES Logistics Modeling Team that allows public health decision makers to rapidly and safely test ideas for improving the delivery of medical supplies.
The researchers convened stakeholders from various backgrounds to explore the implications of vaccine container choices on access, cost, and safety. In a HERMES-generated model of the West African nation of Benin, vial-size decisions had far-reaching and reverberating impacts throughout the vaccine supply chain. For example, increasing vial size reduced total costs by as much as $0.25 per dose—a remarkable impact considering that millions of doses of vaccines are administered globally each year.
“Computational modeling can play a critical role in assessing new technologies and alternative products, whose effects can be difficult or impossible to predict without a tool like HERMES,” says Leila Haidari, HRMES team coordinator, PSC public health applications manager and lead author in the study. “Vaccine supply chains are complex, dynamic systems that can make great use of modeling to inform policy decisions.”
ANSYS PARTNERSHIP RENEWED
PSC’s partnership with Canonsburg, Pa.-based engineering simulation software company ANSYS Inc. has been renewed for another year. Under the continued partnership, ANSYS will provide PSC and its academic users with a no-cost license for ANSYS’s suite of mechanical, computational fluid dynamics and multiphysics software. The software, installed and tested on PSC’s Blacklight supercomputer, will eventually be installed on the upcoming Bridges system (See p.12).
Availability of ANSYS software on PSC’s high performance computing resources has enabled researchers to tackle computationally-intensive problems that otherwise take an inordinate amount of time on a local computer or that require more memory than is available locally. The partnership has also provided ANSYS with supercomputer time for joint software testing between PSC and ANSYS development and application engineering teams.
Under the renewed partnership, academic researchers at U.S. universities who wish to run ANSYS simulations at PSC will continue to be able to obtain time on PSC systems through the NSF XSEDE’s proposal review and allocation process. All universities utilizing these resources must be preauthorized by the partnership group at ANSYS, Inc. All pre- and post-processing must be performed on local computers, and all research must conform to the terms of the license agreement, under which all research must be non-proprietary, public and published/publishable with no restrictions on public availability.