In-network computing (INC) offers orders-of-magnitude performance gains for various applications. However, the existing pipeline-based switch architecture, though well-suited for stateless packet forwarding, imposes three constraints on stateful INC applications: (i) control-plane table management, (ii) scarce memory resources under per-stage layout, and (iii) a strict memory access model. To address these constraints, this paper provides a full-stack solution, covering a programmable chipset Solar-NP, a programming language NPC, and a complete toolchain XuanWu. First, Solar-NP adopts a run-to-completion-based (RTC-based) architecture with three new hardware features: data-plane table management, a hierarchical memory pool, and built-in data structures with atomic access guarantees. Second, to fully exploit the hardware features, our language NPC introduces a new abstraction, namely operation-action table (OAT), that allows table manipulations in the data plane. Finally, our XuanWu toolchain provides various utility software covering the complete workflow of developing, testing, debugging, and deployment. To show the power of our solution, we implement three types of INC cases, including in-network control, in-network telemetry, and in-network storage. Experimental results demonstrate that our solution allows more operations to be offloaded and achieves higher performance than today's pipeline-based solutions.

Period estimation of X-ray pulsars plays an important role in X-ray pulsar based navigation (XPNAV). The fast Lomb periodogram is suitable for period estimation of X-ray pulsars, but its performance in terms of frequency resolution is limited by data length and observation time. Longer observation time or oversampling can be employed to improve frequency analysis results, but with greatly increased computational complexity and large amounts of sampling data. This greatly restricts real-time autonomous navigation based on X-ray pulsars. To resolve this issue, a new method based on frequency subdivision and the continuous Lomb periodogram (CLP) is proposed to improve precision of period estimation using short-time observation data. In the proposed method, an initial frequency is first calculated using fast Lomb periodogram. Then frequency subdivision is performed near the initial frequency to obtain frequencies with higher precision. Finally, a refined period is achieved by calculating the CLP in the obtained frequencies. Real data experiments show that when observation time is shorter than 135 s, the proposed method improves period estimation precision by 1–3 orders of magnitude compared with the fast Lomb periodogram and fast Fourier transform (FFT) methods, with only a slight increase in computational complexity. Furthermore, the proposed method performs better than efsearch (a period estimation method of HEAsoft) with lower computational complexity. The proposed method is suitable for estimating periods of X-ray pulsars and obtaining the rotation period of variable stars and other celestial bodies.