Function-as-a-Service (FaaS) has emerged as a popular cloud paradigm that simplifies software development and deployment by providing scalable and event-driven function execution without the burden of managing servers.FaaS was originally created with a function as a function semantics to enable standalone microservices with adequately short execution time to meet microsecond-scale service-level objectives (SLOs).Unfortunately, today's FaaS systems fundamentally suffer from millisecond-level performance bottlenecks that arise from isolating functions in separate address spaces inside containers or microVMs.Prior work has focused on optimizing FaaS performance, but these systems still fall short of meeting microsecond-level SLOs.In this paper, we present Jord, a FaaS system that revives the original function-as-a-function vision of FaaS.Jord leverages hardware/software co-design to colocate functions in a single address space with user-level in-process memory isolation, extending the capability of traditional virtual memory.By performing memory isolation and management in nanoseconds, Jord enables zero-copy cross-function communication and scalable function dispatch, thereby minimizing FaaS overheads.We demonstrate that Jord can meet microsecond-level SLOs for microservice workloads while performing within 16% of an idealized but insecure baseline and delivering over 2× higher throughput compared to enhanced state-of-the-art systems.

Modern Multi-Processor-Systems-on-Chips (MPSoCs) use Network-on-Chips (NoCs) as a scalable and efficient communication fabric. The applications running on these devices rely on frequent communication with central database servers, which are vulnerable to impersonation attacks by adversarial clones. We propose NoCiPUF, a novel NoC-based intrinsic Physically-Unclonable-Function (PUF) framework for MPSoCs authentication. We re-use the circuit switched nature of NoC with path-pairs as challenges to obtain secret responses, collectively called challenge-response-pairs (CRPs). Due to the random nature of manufacturing variations, equal hop paths exhibit unequal delays. We leverage the delay differences of flits traversing in equal-hop paths to generate unique responses. NoCiPUF is fully-synthesizable and readily scalable as it requires changes only at the behavioral level. To counter Machine-Learning (ML)-based modeling attacks on PUFs, we provide a comprehensive technique and reduce the prediction accuracy to ~52%. NoCiPUF framework incurs low area (0.76%) and power (1.14%) overheads and has no impact on NoC performance in normal mode due to independent authentication mode. Obtained responses have near-ideal PUF metrics and are verified against the NIST randomness test suite. This scheme offers high number of CRPs in larger NoC networks ($>0.74$million CRPs in 5×5 mesh), proving its scalability.