Computer Systems Research
See program guidelines for contact information.
Computers systems support a broad range of applications and technologies that seamlessly integrate with human users. While many key building blocks of computer systems are today commercial technologies, the challenge ahead is to envision new technologies, as well as to combine existing technologies, software, and sensing systems into the computer systems of the future that will span wearable computing, "smart dust," the Internet of Things (IoT), “Smart Cities,” intelligent transportation systems, personalized healthcare, and beyond. Such computer systems will require new, innovative, and visionary approaches to hardware, wired and wireless communications, consideration of human-computer interactions, and new programming languages and compilers that are limited only by the imagination. They will need to be reliable in the presence of unreliable components, adaptive to changing environments, capable of supporting high-throughput applications and large-scale data storage and processing, and able to meet performance and energy objectives for applications ranging from very low-power embedded systems to large high-performance computing systems. Furthermore, computer systems of the future will need to provide mechanisms for ensuring security and privacy.
The Computer Systems Research (CSR) program supports transformative scientific and engineering research leading to the development of the next generation of highly performant, heterogeneous, power-efficient, environmentally sustainable, and secure computer systems. The scope of the program includes embedded and multicore systems and accelerators; mobile and extensible distributed systems; cloud and data-intensive processing systems; and memory, storage, and file systems. The program seeks innovative research proposals that will advance the reliability, performance, power, security and privacy, scalability, and sustainability of computer systems.
CSR proposals should address problems that are appropriate to the CSR Core Area, to one of the current highlighted areas, or to the bridging area of networked systems. Note that proposals that address problems in the CSR highlighted areas are not targeted for special handling or funding -- they simply represent emerging areas or areas of current national interest.
In addition CNS invites proposals that bridge the research areas of CSR and NeTS. Some of the topics specified below in the CSR core program description, along with others, are in the realm of “networked systems,” requiring innovations and expertise in both networking and computer systems. CNS welcomes proposals on these topics, which cross the CSR and NeTS core programs -- and PIs are encouraged to specify proposal titles that begin with “CSR: NeTS:” (see the Proposal Preparation Instructions for details). These proposals will be considered for co-review by the two CNS core programs as appropriate.
CSR proposals are strongly encouraged to include validation plans that describe mechanisms to assess success of the proposed research efforts.
CSR Core Area
The CSR program supports transformative research in computing systems ranging from large-scale systems of tiny sensors and embedded computers to multi-core architectures and operating systems, to mobile and sensor systems, and to warehouse-scale cloud back-end systems and beyond. The essence of the CSR vision is the creation of future environments in which large-scale, complex, and heterogeneous systems seamlessly integrate so that the whole is greater than the sum of the individual parts. These future environments include hardware, software, sensing, and the integration of external factors, such as users and physical environments. In addition, we envision systems that provide high degrees of availability, responsiveness, fault-tolerance, and security, either on demand or by design. Finally, research on novel computer systems that push the limits on system properties including speed, storage and data capacity, power consumption, and real-time requirements is encouraged.
Future computer systems are envisioned that are integrated into everyday activities. These systems include remote, high-performance or embedded devices sensing locally and/or remotely across the surrounding environment, and acting on behalf of individuals and even entire societies. We expect significant systems challenges across an abundance of computing systems, their heterogeneity, and the need to integrate peripheral and smaller with larger computing systems. These and other factors are expected to challenge the way we design, build and sustain computer systems.
The CSR core supports and sustains progress in the contributing disciplinary areas that underlie computing systems including: distributed systems; pervasive and high-performance computing; operating systems and middleware; design and programming models; and real-time, embedded, and hybrid systems.
CSR Highlighted Areas
For this solicitation, there are three CSR highlighted areas: Embedded and Real-time Systems (ERS), Cloud Computing (CC),and Extensible Distributed Systems (EDS). These three areas are described below.
- Embedded and Real-time Systems (ERS)
Embedded and real-time systems control devices and physical or engineered systems that range from hearing aids and pacemakers to automobiles, aircraft, chemical processing plants, electrical power grids, and global aviation infrastructure. The ERS highlight area supports research and education in scientific foundations and technology that will revolutionize the design and development of such systems.
The goal of the ERS highlight area is to supply technologies for designing and building increasingly capable and certifiably dependable embedded and control systems, with real-time, interoperability, survivability, reliability, and security guarantees. Topics of interest include, but are not limited to: embedded systems software and programming methods; real-time services, run-time software and platforms; innovative embedded hardware technology and hardware-software co-design; scalable support for embedded sensing and real-time applications; architecture and design principles for complex embedded systems; and power and resource management and optimization.
- Cloud Computing (CC)
Cloud computing in various forms has rapidly become the dominant method of providing on-demand access to computing infrastructure, data, and software utilities for Internet services. Through the use of virtualization, multi-tenancy, and various service models, even small companies are able to quickly create and scale web and mobile applications.
The main goal of the CC highlight area is to stimulate and promote basic research related to novel theory, algorithms, performance analyses, and applications of cloud computing techniques. Topics of interest include, but are not limited to: novel cloud architectures and storage systems; programming models for the cloud; network support for cloud computing; real-time clouds and quality of service (QoS); cloud management, benchmarking, self-monitoring, prediction, and autonomic resource management; technology trends and sustainable data centers; and cloud service, experiences, economics, and adoption.
Prospective PIs for the CC highlight area are especially encouraged to consider utilizing the two NSFFutureCloud prototypes, Chameleon and CloudLab, when formulating their research plans and submitting CC proposals. Detailed descriptions of the hardware and capabilities of each system are available on their respective websites, www.chameleoncloud.org and www.cloudlab.us. Both systems are currently open for use, with application instructions available on their websites.
- Extensible Distributed Systems (EDS)
A new generation of distributed systems is emerging that operates not simply from the cloud, but extends both to the portable computing devices that are increasingly everywhere around us (in our pockets, on our wrists and ankles, and in our cars and homes, etc.) and indeed also to the individuals who own, operate, and depend upon these systems. Users expect these systems to be robust, reliable, safe, secure, and efficient. At the same time, new applications leveraging these platforms require a rich environment that enables sensing and computing along with communication among the devices and between the devices and warehouse-scale facilities via the cloud. This coupling underpins many ‘smart’ technologies and infrastructures of the future, such as fog computing, smart buildings and informatics infrastructures, intelligent transportation systems, and smart energy distribution and consumption systems, as well as how humans interact with such technologies and infrastructures.
The goal of the EDS highlight area is to support research into the science and design of extensible distributed systems, advancing software and hardware architectures to enable us to integrate and aggregate from the smallest scale systems to systems without bounds. Topics of interest include, but are not limited to: systems models and programming abstractions that make it easy to compose components and modules to create richer applications; architectures that enable humans to better manage and maintain the applications and the systems upon which they run; tools for assessing the safety and security of applications that interact with the physical world; methods and tools for assessing uncertainty and the tradeoffs of uncertainty, time, and usability, as well as for balancing constraints such as power, energy, form factor, tight time, computational capacity, consistency, reliability, dependability, and performance; and approaches for data storage and recovery for EDS.
Both the CSR Core and the CSR highlighted areas seek proposals focused on advances in systems computing and systems programming that are particular to an application domain or a specific hardware platform as well as generic across domains and platforms.
Computer Systems Research (CSR) Staff
Funding Opportunities for the Computer Systems Research (CSR) Program:
Computer and Network Systems (CNS): Core Programs. NSF 15-572
THIS PROGRAM IS PART OF
Computer and Network Systems (CNS): Core Programs