In key-value storage systems, a lot of traffic is concentrated on some of the hotspots. The servers that keep hotspots will become bottlenecks in the whole system because of this skewed workload. Recent studies have demonstrated that load imbalance can be effectively mitigated by deploying a small cache node in front of the back-end servers—commonly referred to as an in-network cache. With the advent of programmable switches, it has become feasible to position this cache node directly on the switch, a critical point through which all traffic naturally flows. A class of in-network cache solutions is proposed to balance workloads among backend servers, but they fall short in supporting dynamic workloads where hotspots vary quickly with time. We show that the bottleneck is caused by the slow rule update speed on switch hardware and programming abstraction — Match-Action Table (MAT). To achieve faster cache update, we designed a new cache calledRAM cachebased on register arrays that support the write-back method. The update of theRAM cacheis done in a data-flow-driven manner, and we ensure its correctness. However, due to the hardware limitations of programmable switches, the RAM cache cannot implement complex cache update policies and has low space utilization. To address workload dynamics without sacrificing system performance, we then design acache hierarchynamed ICH, which combines the advantages of both MAT and register. We carefully devise thecache admission and evictionto keep the cache coherence, continuously available, and the server workload light. Our ICH prototype and extensive experiments demonstrate ICH improves the key-value store throughput for various access patterns (read/write intensive), especially significantly for highly dynamic workloads (up to 155%).

In a key-value storage system, a small amount of hot item will account for most of the traffic. Skewed workloads can lead to load imbalance between servers, and some servers that keep hotspots will become system bottlenecks and impact the performance of the entire system. The recent studies have shown that the load imbalance can be eliminated by placing a small, fast cache node in front of the back-end servers, i.e., in-network cache. And the appearance of programmable switches made it possible to place this node on the switch, the place through which the traffic must pass. While existing in-network cache effectively balances loads between servers in large-scale storage systems, they struggle in write-intensive workloads and lose scalability with increasing client numbers due to imbalanced cache nodes. This paper introduces HarmonyCache, a scalable and high-performance in-network cache system that supports write-back methods. HarmonyCache use cache replication and read-write separation mode, only one node in all the cache is responsible for handling write requests, the other node can only handle read requests. To achieve system scalability and minimize cache coherence overhead, HarmonyCache proposes an adaptive cache replication scheme to decide where the cache should be replicated and how many replications should be made. Additionally, according to the requirements of different cache nodes, we design different types of in-network caches using different switch resources and propose a hybrid cache scheme. Our HarmonyCache prototype and extensive experiments demonstrate substantial improvements in key-value store throughput across various access patterns (read/write intensive), achieving a throughput gain of up to 7.6× over state-of-the-art solutions under skewed write-intensive workloads.

Load imbalance in distributed key-value (KV) storage systems, especially under skewed access patterns across racks or clusters, can significantly degrade performance. We present CacheMigrate, a switch-assisted caching and migration framework that leverages programmable Top-of-Rack (ToR) switches for real-time hotspot detection, in-network caching, and per-key migration state tracking. By decoupling data movement from request handling, CacheMigrate minimizes service disruption and ensures strong consistency during live migration. A distributed migration state management design coordinates ToR and spine switches without centralized bottlenecks. We implement CacheMigrate on a P4-programmable switch and evaluate it using realistic workloads. Results show that CacheMigrate consistently delivers higher throughput and better load balance than existing approaches across diverse workload and system configurations.

Online data-intensive services (OLDI) are subject to stringent service-level objectives (SLOs), yet unstable server response times can result in increased Remote Procedure Calls (RPC) tail latency. Request cloning is an effective technique for mitigating tail latency by masking variations in service time. However, traditional static cloning suffers from performance degradation under high load, while the coordinatorbased dynamic cloning scheme introduces additional delays and scalability challenges. Programmable switches present the opportunity to replicate requests dynamically in the network to eliminate additional delays. Still, they face issues with inconsistent server states and inaccurate load perception, which could lead to server overload and queue head blocking. Moreover, congestion control schemes must be considered due to the replication of requests. To address these challenges, we propose RoClone, which combines real-time status updates and fast-responsive server state awareness with congestion control to alleviate server-side pressure. RoClone overcomes memory and access limitations in programmable switches, ensuring low latency even under high load. We implemented RoClone on a Barefoot Tofino switch. Experimental results demonstrate that RoClone significantly reduces tail latency across various loads and maintains low tail latency under high load. The system reduces the 99th percentile latency by at least 60 % compared to NetClone and by at least 40 % compared to the baseline.

Due to existence of the long along-track baseline (ATB) among the different satellites in a distributed space-based radar (DSBR) system, a large number of grating lobes appear in radar returns, causing the non-continuous detection phenomenon of an air moving target (AMT). To solve this problem, in this paper, a novel approach for ATB distribution design in a DSBR system is proposed. In the proposed algorithm, to reduce the amount of spatial ambiguity points located at the main-lobe region and achieve the best air moving target indication (AMTI), the maximum target detectability ratio (TDR) in the main-lobe region is chosen to be the criteria for the optimal ATB design. The effectiveness of the proposed method is verified by the simulated multi-channel radar data in a DSBR system.

Breakthroughs in Wireless Energy Transfer technologies have revitalized Wireless Rechargeable Sensor Networks. However, how to schedule mobile chargers rationally has been quite a tricky problem. Most of the current work does not consider the variability of scenarios and how many mobile chargers should be scheduled as the most appropriate for each dispatch. At the same time, the focus of most work on the mobile charger scheduling problem has always been on reducing the number of dead nodes, and the most critical metric of network performance, packet arrival rate, is relatively neglected. In this article, we develop a DRL-based Partial Charging algorithm. Based on the number and urgency of charging requests, we classify charging requests into four scenarios. And for each scenario, we design a corresponding request allocation algorithm. Then, a Deep Reinforcement Learning algorithm is employed to train a decision model using environmental information to select which request allocation algorithm is optimal for the current scenario. After the allocation of charging requests is confirmed, to improve the Quality of Service, i.e., the packet arrival rate of the entire network, a partial charging scheduling algorithm is designed to maximize the total charging duration of nodes in the ideal state while ensuring that all charging requests are completed. In addition, we analyze the traffic information of the nodes and use the Analytic Hierarchy Process to determine the importance of the nodes to compensate for the inaccurate estimation of the node’s remaining lifetime in realistic scenarios. Simulation results show that our proposed algorithm outperforms the existing algorithms regarding the number of alive nodes and packet arrival rate.

To achieve tight beam coverage and obtain the optimal target detection performance for a space-based early warning radar (SBEWR) system, the reasonable radar beam design is prerequisite. In this paper, an optimal beam position design algorithm in a SBEWR system is proposed, where the beam position overlap rate (BPOR) choice is reasonably designed. In the proposed algorithm, the maximum signal-to-clutter and noise ratio (SCNR) of the targets located at the beam position edge is chosen as the criterion for the BPOR optimization design. The BPOR designment result provides important guidance and references for beam filling designment and guarantee the effective target detection capability of a SBEWR system.

Due to existence of the long along-track baseline (ATB) between the satellites in a distributed space-based early warning radar (DSBEWR) system, a large number of grating lobes appear in radar returns, causing the noncontinuous detection phenomenon of an aerial moving target (AMT). To solve this problem, in this paper, a novel approach for ATB distribution design in a muti-satellite DSBEWR system is proposed. In the proposed algorithm, to reduce the amount of spatial ambiguity points located at the main-lobe region, the main-lobe gain of the received antenna pattern in the azimuth dimension is designed to be maximum via the non-uniformly intersatellite ATB designment. The effectiveness of the proposed method is verified by the simulated multi-channel radar data in a DSBEWR system.

In this letter, the clutter range dependence issue caused by crab angle is investigated in an airborne multi-channel radar system. A new method is proposed to accomplish the crab angle estimation and the non-stationary error compensation. Firstly, from the center-biased and spectrum-broadened characteristics of the clutter space-time spectrum in the case of non-neglected crab angle, a 1-D cost function based on space-time spectrum broadening degree is constructed to achieve the accurate crab angle estimation. Then, the clutter non-stationary errors can be effectively compensated in the post-Doppler domain, significantly improving the subsequent space time adaptive processing (STAP) performance. Simulation results are presented to validate the feasibility and effectiveness of the proposed method.

Due to the existence of the time synchronization error, the correlation coefficient between the main satellite and auxiliary satellite severely declines, thus deteriorating the clutter suppression ability and aerial moving target indication (AMTI) performance in a distributed space-based early warning radar (DSBEWR) system. In this paper, a novel algorithm is proposed to estimate and calibrate the linear time synchronization error. In the proposed algorithm, the rough linear time synchronization error range is firstly estimated according to the Doppler offset between the same spatial channel in the main satellite and the auxiliary satellite. After obtaining the searched rough error range, the linear time synchronization error can be accurately obtained by finding the max correlation coefficient between the main satellite and auxiliary satellite after performing time synchronization error compensation. The effectiveness of the proposed method is verified by the simulated multi-channel radar data in a DSBEWR system.

Due to existence of the long along-track baseline (ATB) in a distributed space-based early warning radar (DSBEWR) system, a large number of grating lobes appears in both main-lobe and side-lobe clutter regions, causing the discontinuously detectable phenomenon of an aerial moving target (AMT). To solve this problem, in this paper, a novel spatial ambiguity suppression algorithm is proposed. In the proposed algorithm, based on the prior information provided by the system parameters and the signal processing results from the single satellite radar data, after performing the multi-channel clutter rejection, the spatial ambiguity can be eliminated via the joint processing of residual clutter signals corresponding to the single satellite and the muti-satellites. Simulated processing results are provided to validate the effectiveness of the proposed method.

In this letter, a novel method is proposed to suppress the terrain fluctuation clutter based on adaptive broadening processing. In the proposed algorithm, the flat interference phase is first compensated according to the priori radar system parameters. Then, according to the space-time trajectory distribution of clutter edge, the level of clutter Doppler spread in virtue of crab effect is estimated by a cost function related to the clutter eigenvector matrix. Finally, after calculating the clutter suppression weight vector according to the broadened clutter subspace, the non-stationary ground clutter can be robustly rejected. The validity of the proposed method is verified by both the simulated and real-measured multichannel radar data.

Due to the limited size, carrying capacity, power-aperture product, and high hardware cost of satellite platform, the traditional single-platform spaceborne radar system encounters the problems of poor target minimum detectable velocity (MDV) performance, considerably deteriorating the moving target detection performance. To improve the air moving target indication (AMTI) performance, especially for a weak target, distributed space-based radar system (DSBR) becomes a good candidate due to the longer along-track baseline (ATB) and spatial power synthesis. However, due to the sparse configuration of radar baseline distribution, the detection performance of air moving targets (AMTs) will be restricted by many practical factors in an actual DSBR system. In this paper, multi-channel signal models of an observed moving target and ground clutter are accurately established in a DSBR framework, where the error influences of cross-track baseline (CTB), terrain fluctuation, and channel inconsistency response are considered. Then, the influence of the non-ideal factors, including the channel noise, long-intersatellite ATB, long-intersatellite CTB, synchronization errors, and interchannel amplitude and phase inconsistency errors, on the AMTI performance is analyzed term by term. The simulation results provide the useful guidance for the system design of a DSBR with the AMTI tasks.

In this paper, a precise beam position design algorithm for space-based early warning radar (SEWR) based on the actual antenna pattern search is proposed. In the proposed algorithm, the beam positions along elevation and azimuth dimensions are filled according to the main lobe widths of elevation and azimuth antenna patterns, respectively, and thus the beam position designment is more accurate since the range and azimuth angle dependences are considered. Based on the designed beam positions, the precise dwell time can be obtained according to the coverage rata of a SEWR system. The beam position designment results by using the proposed method provide a reference for the beam filling designment of a space-based radar.