UTA researcher aimed at redesign abstractions in virtualized systems to improve efficiency
Credit: UT Arlington
Jia Rao, an assistant professor in the Computer Science and Engineering Department at The University of Texas at Arlington, has been awarded a four-year, $498,000 National Science Foundation Early Career Development, or CAREER, Program grant to redesign abstractions in virtualized systems to improve efficiency.
The Faculty Early Career Development Program is the NSF’s most prestigious award for junior faculty. Winners are outstanding researchers, but also are expected to be outstanding teachers through research, educational excellence and the integration of education and research at their home institutions.
Abstractions are used to hide capacity in computer systems by removing less important details to attend to other, more pertinent ones. One type of abstraction–virtualization–is a key component of cloud computing and has changed how computer systems use resources by allowing multiple virtual computer architectures and systems to run off of a single physical machine. However, performance, cost-effectiveness and predictability issues are keeping virtualization out of domains such as scientific computing and big-data analytics.
“The problem with existing abstractions in virtual systems is that, despite the benefits to cloud computing, abstractions incur a lot of inefficiency and unpredictability to cloud users,” Rao said. “Our intention is to improve resource management in any kind of virtual system to enable elastic, effective and efficient use of those resources.”
Virtualization is the process of creating a virtual copy of real physical resources to help ease the management of computer systems and allow maximum flexibility in resource management. Using virtualization, it is possible to build a computer system where several virtual machines, each assigned to different users, can run off one physical machine. However, it is difficult to meet each individual’s needs and maintain high utilization and efficiency in the system due to semantic gaps, or critical missing information between levels of abstraction.
For instance, a user of a virtual machine has the illusion of a dedicated resource and continuous availability, but those resources are actually built upon a physical resource shared among many users. Since resource management software was designed for a physical system with continuous availability, the software doesn’t work efficiently in a virtualized environment because each user’s demands on the physical system are different.
“We hope to bridge these semantic gaps by augmenting existing, regularly adopted abstractions while retaining the benefits of abstraction, including modality, security and portability,” Rao said. “We will then use the knowledge we develop in designing abstractions in virtualized systems to guide the design of abstractions in future hardware systems that will support multi-tenacy.”
His research is an example of data-driven discovery, one of the themes of UTA’s Strategic Plan 2020: Bold Solutions | Global Impact, said Peter Crouch, dean of the College of Engineering.
“Cloud computing has changed how the world uses computer networks, and Dr. Rao’s research will make the cloud more efficient. This could lead to even greater opportunities for business, the scientific community and anyone who uses the cloud as part of their daily lives,” Crouch said.
Seven other UTA faculty have active NSF CAREER Award support:
- Matthew Walsh of the Biology Department received $600,000 in 2017 to study whether behavioral plasticity promotes or constrains adaptation.
- Majie Fan of the Earth and Environmental Sciences Department received $485,627 in 2015 to enhance understanding of how the Rocky Mountains and its modern, elevated landscape came to be.
- Yi Hong in the Bioengineering Department received $500,000 in 2016 to develop a polymer that will allow engineers to develop a scaffold that is flexible, conductive and biodegradable for biomedical applications such as tissue repair.
- Junzhou Huang of the Computer Science and Engineering Department received $545,763 in 2016 to discover a process by which image-omics data can be combined into files that are small enough that current computing technology will allow scientists to better predict how long a patient will live and how best to treat him or her.
- Ankur Jain in the Mechanical and Aerospace Engineering Department received $500,000 in 2016 to develop a fundamental understanding of how heat flows in materials within a Li-ion battery so they can be used safely in more applications.
- Alice Sun in the Electrical Engineering Department received $500,000 in 2016 to develop an all-liquid optofluidic laser that could better detect cancer in the comfort of a doctor’s office.
- Kyungsuk Yum in the Materials Science and Engineering Department received $500,000 to design and develop bioinspired 3D materials with programmed shapes and motions.
The Computer Science and Engineering Department offers degrees in computer engineering, computer science and software engineering. The Computer Engineering program was ranked No. 80 and the Computer Science program No. 90 in the U.S. News and World Report 2019 graduate rankings. The department’s internationally recognized faculty members are engaged in breakthrough research across the leading areas of big data and large-scale computing, biocomputing and health informatics, computer networks, computer vision and multimedia, database and information systems, embedded systems and mobile computing, machine learning and data mining, robotics and artificial intelligence, security and privacy, software engineering and sustainable computing.