Why Rethink the OS?
Operating Systems (OSes) facilitate the simultaneous operation of multiple applications by managing a device's hardware, but they rely today on outdated manually crafted heuristic policies that cannot adapt to modern computing advancements. These policies struggle to efficiently support new hardware technologies and the evolving and dynamic demands of applications across platforms with significant inherent complexity like the cloud, personal robots, and mobile access edges. The continuous and manual updating of OSes to match technological progress is costly, time-consuming, and increasingly untenable in the face of rapid innovation.
Vision
This project aims to build an intelligent, self-adaptive Operating System (OS) that can optimally support modern applications’ performance and resource needs in diverse scenarios that are highly dynamic and exhibit significant complexity.
In our clean-slate Learning Directed OS, or LDOS, advanced machine learning, not human-crafted heuristics, makes rich data-driven resource management decisions that meet application needs at maximal efficiency, avoiding overprovisioning. Further, a single base LDOS implementation can auto-adapt to broad and different settings, avoiding manual heuristic design and customization.
Benefits of LDOS
LDOS can fundamentally change the direction of computing and bring fundamental benefits along many axes. LDOS can enable computing devices and infrastructure to operate at high efficiency, reducing cost and energy use. LDOS enables developers to create novel real-time applications with complex resource needs. Users will enjoy LDOS-enabled low-cost devices and transformative apps. LDOS simplifies systems management making systems "self-driving" via the automation of labor-intensive OS implementation and tuning tasks.
Transformative Use Cases
Our project aims to show the transformative real-world impact of LDOS via three compelling use cases:
Autonomous Mobile Service Robots that concurrently run novel 3rd party apps extending the robot's core functionality.
Assisting humans in day-to-day activities, similar to smartphones today
Self-Managing Clouds that achieve high dependability and high utilization.
Significantly lowering computing’s global ecological impact
Real-time 6G Mobile Access Edges that enable novel real-time edge applications.
Unlocking domains like smart cities and smart factories
Education and Broadening Participation
LDOS, with its one foot in popular Machine Learning, offers a unique opportunity to reboot excitement around computer systems.
Our Expedition will use LDOS as the basis to create new modules and courses highlighting the beautiful interplay of systems and machine learning principles and develop novel certificates and MS programs.
We are developing new initiatives for broadening participation that cultivate leadership among underrepresented groups in ML and computer systems by offering LDOS-inspired tailored programs at different educational levels from middle school to higher education.
Collaboration with OS Practitioners
We are partnering closely with leading industry giants to develop LDOS, evaluate and demonstrate it in realistic deployments, create an open-source ecosystem, and encourage broad adoption across the computing industry.
