$μ$TAS: Design and implementation of Time Aware Shaper on SmartNICs to achieve bounded latency
Authors:
Joydeep Pal,
Deepak Choudhary,
Nithish Krishnabharathi Gnani,
Chandramani Singh,
T. V. Prabhakar
Abstract:
Time-Aware Shaper (TAS) is a time-triggered scheduling mechanism that ensures bounded latency for time-critical Scheduled Traffic (ST) flows. The Linux kernel implementation (a.k.a TAPRIO) has limited capabilities due to varying CPU workloads and thus does not offer tight latency bound for the ST flows. Also, currently only higher cycle times are possible. Other software implementations are limite…
▽ More
Time-Aware Shaper (TAS) is a time-triggered scheduling mechanism that ensures bounded latency for time-critical Scheduled Traffic (ST) flows. The Linux kernel implementation (a.k.a TAPRIO) has limited capabilities due to varying CPU workloads and thus does not offer tight latency bound for the ST flows. Also, currently only higher cycle times are possible. Other software implementations are limited to simulation studies without physical implementation. In this paper, we present $μ$TAS, a MicroC-based hardware implementation of TAS onto a programmable SmartNIC. $μ$TAS takes advantage of the parallel-processing architecture of the SmartNIC to configure the scheduling behaviour of its queues at runtime. To demonstrate the effectiveness of $μ$TAS, we built a Time-Sensitive Networking (TSN) testbed from scratch. This consists of multiple end-hosts capable of generating ST and Best Effort (BE) flows and TSN switches equipped with SmartNICs running $μ$TAS. Time synchronization is maintained between the switches and hosts. Our experiments demonstrate that the ST flows experience a bounded latency of the order of tens of microseconds.
△ Less
Submitted 11 October, 2023;
originally announced October 2023.
EdgeP4: A P4-Programmable Edge Intelligent Ethernet Switch for Tactile Cyber-Physical Systems
Authors:
Nithish Krishnabharathi Gnani,
Joydeep Pal,
Deepak Choudhary,
Himanshu Verma,
Soumya Kanta Rana,
Kaushal Mhapsekar,
T. V. Prabhakar,
Chandramani Singh
Abstract:
Tactile Internet based operations, e.g., telesurgery, rely on end-to-end closed loop control for accuracy and corrections. The feedback and control are subject to network latency and loss. We design two edge intelligence algorithms hosted at P4 programmable end switches. These algorithms locally compute and command corrective signals, thereby dispense the feedback signals from traversing the netwo…
▽ More
Tactile Internet based operations, e.g., telesurgery, rely on end-to-end closed loop control for accuracy and corrections. The feedback and control are subject to network latency and loss. We design two edge intelligence algorithms hosted at P4 programmable end switches. These algorithms locally compute and command corrective signals, thereby dispense the feedback signals from traversing the network to the other ends and save on control loop latency and network load. We implement these algorithms entirely on data plane on Netronome Agilio SmartNICs using P4. Our first algorithm, $\textit{pose correction}$, is placed at the edge switch connected to an industrial robot grip** a tool. The round trip between transmitting force sensor array readings to the edge switch and receiving correct tip coordinates at the robot is shown to be less than $100~μs$. The second algorithm, $\textit{tremor suppression}$, is placed at the edge switch connected to the human operator. It suppresses physiological tremors of amplitudes smaller than $100~μm$ which not only improves the application's performance but also reduces the network load up to $99.9\%$. Our solution allows edge intelligence modules to seamlessly switch between the algorithms based on the tasks being executed at the end hosts.
△ Less
Submitted 19 September, 2023;
originally announced September 2023.
