Computing in the network is a new research area that has emerged over the past few years. In-network computing refers to the execution of programs typically running on end-hosts within network devices. The computation in this model is done within the network, using devices that have already been deployed and are already used to forward network traffic. In contrast to traditional network computing where the computation is done by computers deplyoed in a network, the computations are performed by programmable switches. The two key benefits of in-network computing are high throughput and low latency. Nowadays, switch ASICs process up to ten billion packets per second, supporting billions of operations per second per offloaded application. These switches are designed as pipelines, working on streams of packets without stalls. In general, they support sub-microsecond latency with low variance in non-overloaded situations. In in-network computing transactions are terminated within their network path rather than reaching an end-host (e.g. a server). In this way, the latency introduced by the end-host can be saved, significantly reducing the response time. It has recently been demonstrated by existing applications that in many scenarios 10 000 times higher performance can be reached with performing computation in the network compared to their server-based counterparts.

With the advent of programmable data planes [1,2], a new era has begun in computer networking, transforming networks from packet forwarding infrastructures to programmable end-to-end platforms. The main idea of Software-Defined Networking 1.0 (SDN 1.0) was the separation of data and control planes in switching devices and opening the control plane supervised by a logically centralized software controller. In addition to the control plane programmability of SDN 1.0, data plane programmability allows network operators to get full control over their infrastructure. They cannot only describe how the control plane applications need to fill tables in the packet forwarding devices, but they can define how these devices should handle or process packets. In contrast to SDN 1.0, modifying the low-level packet processing does not require redesigning the underlying switching microchip (ASIC); it can easily be described in software. As Nick McKeown said in his keynote at NetDev 0x14: „We will no longer think in terms of protocols. Instead, we will think in terms of software. All functions and “protocols” will have migrated up and out of hardware into software. Throughout the Internet.”. In future programmable networks, the control plane can continuously monitor the network states and react to them in many ways. This extended visibility will lay down the foundations of self-driving and intent-based networks. Deeply programmable networks [3] will reshape computer networks in the future and pave the road towards unforeseen innovations in the field. However, they will also result in many challenges relating to security, reliability, and trust in the networking software. Our research focuses on these challenges, aiming to answer different research questions from the following fields: * New network algorithms for intent-based traffic engineering solely in the data plane * Programming network as a big switch, disaggregation of data and control plane programs * Utilization-aware routing * In-network acceleration of various applications * Anomalous and malicious traffic behavior detection and mitigation in the data plane * Detection and diagnosis of performance problems by measurements in the data plane * Fast recovery after failure detection References: [1] Pat Bosshart, Dan Daly, Glen Gibb, Martin Izzard, Nick McKeown, Jennifer Rexford, Cole Schlesinger, Dan Talayco, Amin Vahdat, George Varghese, David Walker, P4: Programming Protocol-Independent Packet Processors, ACM Sigcomm Computer Communications Review (CCR). Volume 44, Issue #3 (July 2014) [2] P4 Consortium: http://p4.org

Despite extensive research and standardization in the area of Quality of Service (QoS), most of the developed solutions have not been deployed in practice. Proponents of overprovisioning argue that it is much easier and more efficient to add capacity when needed than to build and maintain complex QoS mechanisms that only provide minor improvement during congestion. Network congestion together with the ways of avoiding its impacts on the end users is a recurrent aspect along time, especially nowadays where the network is becoming a critical societal asset. Recently, the unfortunate pandemic situation due to COVID-19 made evident this critical fact, revealing the need of defining proper mechanisms for improving its robustness and ensuring sufficient quality of experience for end users.

We work on a core-stateless resource sharing framework called Per Packet Value. More details are available at http://ppv.elte.hu

The evolution towards sixth-generation (6G) mobile networks is transforming communication infrastructures into deeply programmable, cloud-native, and AI-driven platforms. Beyond higher data rates and lower latency, 6G networks must support heterogeneous applications, including immersive extended reality (XR), autonomous systems, industrial automation, and distributed artificial intelligence. These applications introduce diverse requirements for performance, reliability, energy efficiency, and computing resources, calling for intelligent networks that dynamically adapt to application demands and changing operating conditions [1,2]. Our research contributes to this transformation by developing programmable network architectures, intelligent resource management mechanisms, and application-aware communication solutions. Previous results include hybrid hardware/software architectures for accelerating 5G network functions [3], frameworks for composing and disaggregating programmable data-plane functions [4], and scalable Quality of Service (QoS) and Quality of Experience (QoE) mechanisms [5,6]. We have also investigated digital-twin-assisted radio resource management [7], machine learning for cellular network analytics [8], and energy-efficient resource optimization [9]. Our long-term vision is to enable autonomous, application-aware 6G networks that jointly optimize communication and computing resources according to application-level objectives. Such networks should continuously monitor their operating conditions, understand application requirements, and dynamically adapt network behavior to maximize performance and user experience while ensuring reliability and energy efficiency.

Our research focuses on the following directions:

Deeply programmable 6G architectures: Flexible network architectures integrating programmable data planes, cloud-native network functions, and distributed control mechanisms [1–4].AI-native network management: Intelligent and autonomous mechanisms for monitoring, performance prediction, resource optimization, and closed-loop network orchestration [7,8].Application-aware QoS and QoE optimization: Novel scheduling, traffic management, and resource-sharing algorithms considering packet importance, application requirements, and user-perceived quality [5,6].Joint communication and computing optimization: Dynamic resource allocation and service placement across radio access, transport, edge, and cloud infrastructures, considering performance and energy efficiency [2,9].Goal-oriented and semantic-aware communication: Communication mechanisms that optimize the utility and relevance of information according to application-level objectives.Reliable and trustworthy autonomous networks: Security, resilience, and verification mechanisms ensuring dependable operation of programmable and AI-controlled 6G infrastructures.

The ultimate goal is to establish the foundations of intelligent, sustainable, and trustworthy 6G networks in which communication and computing resources dynamically cooperate to fulfill application-level goals.

[1] G. Pongrácz, A. Mihály, I. Gódor, S. Laki, A. Nanos, C. Papagianni (2023). Towards Extreme Network KPIs with Programmability in 6G. ACM MobiHoc, pp. 340–345. https://doi.org/10.1145/3565287.3617610

[2] F. Paolucci et al. (2026). End-to-End Latency Assurance for Distributed Augmented Reality over Programmable 6G Networks: A DESIRE6G Demonstration. Computer Networks, Vol. 286, 112443. https://doi.org/10.1016/j.comnet.2026.112443

[3] S. K. Singh et al. (2023). Hybrid P4 Programmable Pipelines for 5G gNodeB and User Plane Functions. IEEE Transactions on Mobile Computing, Vol. 22, No. 12, pp. 6921–6937. https://doi.org/10.1109/TMC.2022.3201512

[4] F. Brisch, A. Kassler, S. Laki, P. Hudoba (2024). P4-MTAGG – a Framework for Multi-Tenant P4 Network Devices. IEEE CNSM. https://doi.org/10.23919/CNSM62983.2024.10814635

[5] F. Fejes, Sz. Nádas, G. Gombos, S. Laki (2022). DeepQoS: Core-Stateless Hierarchical QoS in Programmable Switches. IEEE Transactions on Network and Service Management. https://doi.org/10.1109/TNSM.2022.3152017

[6] M. Fekete, G. Gombos, A. Mihály, Sz. Nádas, S. Laki (2026). Towards QoE-Aware Core-Stateless Resource Sharing. ACM CoNEXT EMS 2026 Workshop, accepted paper.

[7] Cs. Györgyi, J. Pető, P. Vörös, G. Szabó, S. Laki (2023). Radio Propagation Digital Twin Aided Multi-Point Transmission With In-Network Dynamic On-Off Switching. IEEE Access, Vol. 11, pp. 128843–128858. https://doi.org/10.1109/ACCESS.2023.3333525

[8] K. Hajdú-Szűcs et al. (2025). Ensemble Graph Attention Networks for Cellular Network Analytics: From Model Creation to Explainability. IEEE Transactions on Network and Service Management. https://doi.org/10.1109/TNSM.2024.3436677

[9] K. Kecskeméti, J. A. Brito, G. Gombos, S. Laki, J. I. Moreno, L. M. Contreras (2026). Energy-Efficient Resource Management Optimization Using Programmable Switches. IEEE ICC 2026. https://doi.org/10.1109/ICC59461.2026.11587383