Federico De Marchi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Federico De Marchi has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Yiting Xia
Max Planck Institute for Informatics
8 shared publications - Jialong Li
Shenzhen University of Advanced Technology, Shenzhen, China
7 shared publications - Wei Bai
NVIDIA
5 shared publications - Yiming Lei
Max Planck Institute for Informatics
3 shared publications - Aoyu Gong
EPFL
2 shared publications - Balakrishnan Chandrasekaran
Vrije Universiteit Amsterdam, Amsterdam, Netherlands
2 shared publications
Latest Publications
- Open Access
- 30 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5-6
Poster: Dynamic Path Allocation in Flat Topologies with Pulse
- 30 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5-6
Open AccessPoster: Dynamic Path Allocation in Flat Topologies with Pulse
- Open Access
- 27 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 842-860
Unlocking Superior Performance in Reconfigurable Data Center Networks with Credit-Based Transport
- 27 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 842-860
Open AccessUnlocking Superior Performance in Reconfigurable Data Center Networks with Credit-Based Transport
- 7 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 33-48
Unlocking Diversity of Fast-Switched Optical Data Center Networks With Unified Routing
Optical data center networks (DCNs) are emerging as a promising solution for cloud infrastructure in the post-Moore’s Law era, particularly with the advent of “fast-switched” optical architectures capable of circuit reconfiguration at microsecond or even nanosecond scales. However, frequent reconfiguration of optical circuits introduces a unique challenge: in-flight packets risk loss during these transitions, hindering the deployment of many mature optical hardware designs due to the lack of suitable routing solutions. In this paper, we present Unified Routing for Optical networks (URO), a general routing framework designed to support fast-switched optical DCNs across various hardware architectures. URO combines theoretical modeling of this novel routing problem with practical implementation on programmable switches, enabling precise, time-based packet transmission. Our prototype on Intel Tofino2 switches achieves a minimum circuit duration of $\mathrm {2~\mu \text {s} }$ , ensuring end-to-end, loss-free application performance. Large-scale simulations using production DCN traffic validate URO’s generality across different hardware configurations, demonstrating its effectiveness and efficient system resource utilization.
- 7 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 33-48
Unlocking Diversity of Fast-Switched Optical Data Center Networks With Unified Routing
Optical data center networks (DCNs) are emerging as a promising solution for cloud infrastructure in the post-Moore’s Law era, particularly with the advent of “fast-switched” optical architectures capable of circuit reconfiguration at microsecond or even nanosecond scales. However, frequent reconfiguration of optical circuits introduces a unique challenge: in-flight packets risk loss during these transitions, hindering the deployment of many mature optical hardware designs due to the lack of suitable routing solutions. In this paper, we present Unified Routing for Optical networks (URO), a general routing framework designed to support fast-switched optical DCNs across various hardware architectures. URO combines theoretical modeling of this novel routing problem with practical implementation on programmable switches, enabling precise, time-based packet transmission. Our prototype on Intel Tofino2 switches achieves a minimum circuit duration of $\mathrm {2~\mu \text {s} }$ , ensuring end-to-end, loss-free application performance. Large-scale simulations using production DCN traffic validate URO’s generality across different hardware configurations, demonstrating its effectiveness and efficient system resource utilization.
- Open Access
- 4 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 4-6
POSTER: Opportunistic Credit-Based Transport for Reconfigurable Data Center Networks with Tidal
The slowdown of Moore's Law for merchant silicon has motivated the adoption of optical circuit switches (OCSes) in data center networks (DCNs). As illustrated in Fig. 1, reconfigurable data center networks (RDCNs) have thus been developed to dynamically allocate high-bandwidth circuits between Top-of-Rack switches (ToRs) through OCSes [5, 6, 8, 17, 18]. Thanks to the increasing reconfiguration speed of OCSes, modern RDCNs offer both high throughput for bulk traffic "elephant" flows and low latency for latency-sensitive "mice" flows with rapid circuit reconfigurations [3, 9, 13, 14, 16].
- 4 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 4-6
Open AccessPOSTER: Opportunistic Credit-Based Transport for Reconfigurable Data Center Networks with Tidal
The slowdown of Moore's Law for merchant silicon has motivated the adoption of optical circuit switches (OCSes) in data center networks (DCNs). As illustrated in Fig. 1, reconfigurable data center networks (RDCNs) have thus been developed to dynamically allocate high-bandwidth circuits between Top-of-Rack switches (ToRs) through OCSes [5, 6, 8, 17, 18]. Thanks to the increasing reconfiguration speed of OCSes, modern RDCNs offer both high throughput for bulk traffic "elephant" flows and low latency for latency-sensitive "mice" flows with rapid circuit reconfigurations [3, 9, 13, 14, 16].
- Open Access
- 4 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 86-88
DEMO: An Open Research Framework for Optical Data Center Networks
Optical data center networks (DCNs) have emerged as a promising design for cloud network infrastructure. However, research in this field is constrained by the requirement for specialized hardware and software stacks and the high engineering barriers of building optical testbeds. To address these challenges, we present an experimental platform that can realize diverse optical DCN architectures in a plug-and-play manner. We demonstrate the ease of realizing different architectures with Python scripts and show performance comparisons of running end-to-end cloud applications.
- 4 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 86-88
Open AccessDEMO: An Open Research Framework for Optical Data Center Networks
Optical data center networks (DCNs) have emerged as a promising design for cloud network infrastructure. However, research in this field is constrained by the requirement for specialized hardware and software stacks and the high engineering barriers of building optical testbeds. To address these challenges, we present an experimental platform that can realize diverse optical DCN architectures in a plug-and-play manner. We demonstrate the ease of realizing different architectures with Python scripts and show performance comparisons of running end-to-end cloud applications.
- Open Access
- 4 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 433-448
Uniform-Cost Multi-Path Routing for Reconfigurable Data Center Networks
Reconfigurable data center networks (RDCNs) are arising as a promising data center network (DCN) design in the post-Moore's law era. However, the constantly reconfigured network topology in RDCNs invalidates the assumption of using hop count as the cost metric for routing, e.g., the status quo Equal-Cost Multi-Path routing (ECMP) in traditional DCNs. Unfortunately, existing routing solutions in RDCNs stick to the old assumption and deliver suboptimal performance either high in latency or low in bandwidth efficiency. In this paper, we redefine the cost metric for RDCN routing with uniform cost to unify the effects of topology disruption and hop count on latency and bandwidth efficiency. We propose Uniform-Cost Multi-Path routing (UCMP), an ECMP equivalent for RDCNs, where minimizing uniform cost leads flows of various sizes to the right balance between latency and bandwidth efficiency. Our simulation shows that UCMP achieves 53% to 98% lower flow completion time (FCT) and 1.55× bandwidth efficiency compared to the state-of-the-art RDCN routing strategy, and our testbed implementation demonstrates sustainable switch resource usage of UCMP as RDCNs scale.
- 4 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 433-448
Open AccessUniform-Cost Multi-Path Routing for Reconfigurable Data Center Networks
Reconfigurable data center networks (RDCNs) are arising as a promising data center network (DCN) design in the post-Moore's law era. However, the constantly reconfigured network topology in RDCNs invalidates the assumption of using hop count as the cost metric for routing, e.g., the status quo Equal-Cost Multi-Path routing (ECMP) in traditional DCNs. Unfortunately, existing routing solutions in RDCNs stick to the old assumption and deliver suboptimal performance either high in latency or low in bandwidth efficiency. In this paper, we redefine the cost metric for RDCN routing with uniform cost to unify the effects of topology disruption and hop count on latency and bandwidth efficiency. We propose Uniform-Cost Multi-Path routing (UCMP), an ECMP equivalent for RDCNs, where minimizing uniform cost leads flows of various sizes to the right balance between latency and bandwidth efficiency. Our simulation shows that UCMP achieves 53% to 98% lower flow completion time (FCT) and 1.55× bandwidth efficiency compared to the state-of-the-art RDCN routing strategy, and our testbed implementation demonstrates sustainable switch resource usage of UCMP as RDCNs scale.
- Open Access
- 4 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 7-13
Rethinking Transport Protocols for Reconfigurable Data Centers: An Empirical Study
Fast-reconfiguring data center network architectures are rising as an answer to the expanding gap between traditional switch capacities and cloud traffic demands. These networks feature frequently changing connections and dynamic routing schemes, creating new challenges for the transport layer. We find that current available schemes have deficiencies when running on top of reconfigurable network fabrics, hindering their performance. Recent customized solutions are also unable to perform optimally when faced with the high number of topology configurations and routing strategies a reconfigurable network can boast. We conduct an empirical analysis of representative transport schemes for these architectures, which will provide insight for future transport designs.
- 4 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 7-13
Open AccessRethinking Transport Protocols for Reconfigurable Data Centers: An Empirical Study
Fast-reconfiguring data center network architectures are rising as an answer to the expanding gap between traditional switch capacities and cloud traffic demands. These networks feature frequently changing connections and dynamic routing schemes, creating new challenges for the transport layer. We find that current available schemes have deficiencies when running on top of reconfigurable network fabrics, hindering their performance. Recent customized solutions are also unable to perform optimally when faced with the high number of topology configurations and routing strategies a reconfigurable network can boast. We conduct an empirical analysis of representative transport schemes for these architectures, which will provide insight for future transport designs.