Time-Aware Applications, Computers, and Communication Systems (TAACCS)

Marc A. Weiss, John C. Eidson, Charles Barry, David Broman, Bob Iannucci, Edward A. Lee, Kevin Stanton, Leon Goldin2015 · DOI

Abstract

A new economy built on the massive growth of endpoints on the internet will require precise and verifiable timing in ways that current systems do not support. Applications, computers, and communications systems have been developed with modules and layers that optimize data processing but degrade accurate timing. State-of-the-art systems now use timing only as a performance metric. Correctness of timing as a metric cannot currently be designed into systems independent of hardware and/or software implementations. To enable the massive growth predicted, accurate timing needs cross-disciplinary research to be integrated into these existing systems. This paper reviews the state of the art in six crucial areas central to the use of timing signals in these systems. Each area is shown to have critical issues requiring accuracy or integrity levels of timing, that need research contributions from a range of disciplines to solve.

Introduction: The Shift from Performance to Correctness

The rapid expansion of the "Internet of Everything" (IoE) promises a future where billions of devices—from autonomous vehicles to smart grid sensors—interact in real-time. However, this vision faces a fundamental architectural hurdle: our current computing and networking stacks were never designed to handle time as a first-class citizen. In modern systems, timing is treated as a side effect of performance rather than a fundamental correctness criterion.

While the asynchronous nature of the internet has allowed for incredible flexibility and scalability, it is increasingly inadequate for Cyber-Physical Systems (CPS). In these environments, a data packet that arrives late or out of sequence isn't just a performance lag; it represents a logical failure of the system. To address this, the National Institute of Standards and Technology (NIST) proposes the Time-Aware Applications, Computers, and Communication Systems (TAACCS) framework—a multi-layered research roadmap designed to integrate precise, verifiable timing across the entire technological stack.

The TAACCS Research Framework

The TAACCS model categorizes the challenges of time-aware systems into six distinct but interdependent layers. Addressing the gaps in these layers requires a cross-disciplinary approach involving physics, electrical engineering, computer science, and network theory:

  1. Oscillators and Clocks: The hardware foundations of frequency and time-keeping. Research aims to improve stability and reduce the cost of high-precision components like Chip Scale Atomic Clocks (CSACs).
  2. Time and Frequency Transfer: Protocols for synchronizing clocks across distances. While NTP is ubiquitous, it lacks the precision required for modern CPS, necessitating more robust protocols like the Precision Time Protocol (PTP).
  3. Time-Aware Networks: Hardware such as bridges and routers that must be capable of controlling latency and jitter to ensure deterministic delivery of time-sensitive data.
  4. Timing Support for Applications: Hardware and software architectures, such as Precision Timed (PRET) machines, that allow for predictable execution times regardless of system load.
  5. Development Environments: Tools for simulating and verifying timing properties during the design phase, moving away from "trial and error" timing adjustments.
  6. Time-Aware Applications: High-level systems that exploit precise timing, such as Google Spanner or distributed power grid management.

The "Asynchrony" Problem in Modern Computing

The success of the OSI (Open Systems Interconnection) model lies in its ability to abstract layers, allowing developers to write software without worrying about the underlying physical hardware. However, this abstraction has created a "blind spot" for timing. Each layer in the stack introduces non-deterministic delays—jitter—that accumulate as a signal moves from the physical layer to the application.

This lack of temporal awareness leads to critical failures in high-stakes environments. In financial markets, poor synchronization can result in "non-causal" transactions, where a trade appears to be settled before it was even initiated. In industrial automation, the lack of a shared global clock forces components to use inefficient "wait states," where hardware sits idle to ensure data is ready, rather than operating on a precisely coordinated schedule.

State of the Art and Technical Benchmarks

The TAACCS review provides a quantitative snapshot of the current landscape of timing technology, highlighting the gap between standard consumer hardware and specialized research implementations.

Oscillator Performance

Hardware selection involves a trade-off between cost, power, and accuracy. Standard Quartz crystals (XO) are inexpensive but offer low accuracy (approximately $10^{-5}$). In contrast, Chip Scale Atomic Clocks (CSACs) provide significantly higher accuracy ($10^{-10}$) in a low-power, small form factor, though their current price point (~$1500) remains a barrier for mass-market IoT adoption.

Network Synchronization

While standard NTP is typically limited to millisecond-level accuracy, specialized implementations of PTP are pushing the boundaries of what is possible over fiber optics. Projects like CERN’s "White Rabbit" have demonstrated the ability to achieve sub-nanosecond accuracy (100 ps) and extremely high precision (8 ps) over distances of 5 km. However, achieving this level of performance in heterogeneous, wide-area networks remains a significant challenge due to varying delays across different router ports (the "multicast diameter" problem).

Security and Resilience

A major concern highlighted in the TAACCS framework is the vulnerability of our current timing infrastructure. Much of the world's critical infrastructure relies on Global Navigation Satellite Systems (GNSS), such as GPS, for timing. These signals are weak and vulnerable to both accidental interference and malicious spoofing. The paper emphasizes the urgent need for network-based backups like IEEE 1588 (PTP) and local "holdover" capabilities using CSACs to maintain accuracy when external signals are lost.

Conclusion and Future Directions

Transitioning to a time-aware world requires a fundamental shift in how we think about system design. Future research must focus on scalability for trillion-device networks, robust security for timing protocols, and the development of programming languages where time is a first-class citizen. By treating timing as a correctness criterion, we can unlock the next generation of distributed systems, ensuring that the Internet of Everything is as reliable as it is connected.

Summary updated

Abstract

A new economy built on the massive growth of endpoints on the internet will require precise and verifiable timing in ways that current systems do not support. Applications, computers, and communications systems have been developed with modules and layers that optimize data processing but degrade accurate timing. State-of-the-art systems now use timing only as a performance metric. Correctness of timing as a metric cannot currently be designed into systems independent of hardware and/or software implementations. To enable the massive growth predicted, accurate timing needs cross-disciplinary research to be integrated into these existing systems. This paper reviews the state of the art in six crucial areas central to the use of timing signals in these systems. Each area is shown to have critical issues requiring accuracy or integrity levels of timing, that need research contributions from a range of disciplines to solve.
BibTeX
@misc{weiss2015time,
  author = {Marc A. Weiss and John C. Eidson and Charles Barry and David Broman and Bob Iannucci and Edward A. Lee and Kevin Stanton and Leon Goldin},
  title = {{Time-Aware Applications, Computers, and Communication Systems (TAACCS)}},
  year = {2015},
  doi = {10.6028/nist.tn.1867},
}
Figure