Dirk Kutscher

Our paper on "MD2G-Cast: Relay-Coordinated Multicast for Scalable
Volumetric Streaming over MoQ" has been accepted for publication at ACM Multimedia 2026..

Abstract

Volumetric streaming for immersive 6DoF experiences remains challenging to scale in multi-user settings. Existing systems treat overlapping field-of-view content as independent per-user flows, leading to redundant data replication across the network. This redundancy saturates shared links and manifests as unstable 𝑃99 tail latency, directly degrading interactive continuity.

We present MD2G-Cast, a relay-coordinated multicast framework over Media over QUIC (MoQ). Our key insight is that relays can act as decision points rather than passive forwarders when volumetric signals, such as viewing overlap, device heterogeneity, and bandwidth dynamics, are incorporated into relay decisions. MD2G-Cast enables relays to form reusable multicast groups and jointly schedule layered delivery, ensuring shared base-layer content is delivered once while enhancement layers are selectively assigned.

We implement MD2G-Cast on top of MoQ and evaluate it on a trace-driven testbed with real-world bandwidth and FoV traces. MD2G-Cast improves average QoE by 19% over multicast baselines while reducing 𝑃99 steady-state delivery delay by up to 30%. Under high concurrency, it achieves 53% higher QoE and up to 58.8% lower 𝑃99 steady-state delivery delay compared to a unicast baseline.

Reference

Ruonan Chai, Yisu Wang, Zili Meng, Dirk Kutscher; MD2G-Cast: Relay-Coordinated Multicast for Scalable Volumetric Streaming over MoQ, accepted for publication at ACM Multimedia 2026; Rio de Janeiro, Brazil — 10–14 November 2026

Our paper on "MultiMoQ: Multi-Access Media-Over-QUIC for Robust Immersive Video Streaming" has been accepted for publication at ACM Multimedia 2026.

Abstract

Live immersive video streaming, particularly 360° video, is increasingly adopted in applications such as virtual events, sports broadcasting, and remote education. Existing approaches struggle to support high-bitrate immersive streaming for large numbers of concurrent users, with coarse-grained delivery limiting responsiveness and insufficient support for coordinating concurrent tile streams. Media over QUIC (MoQ) has recently emerged as a promising solution for large-scale media delivery, yet it lacks robustness under bandwidth-constrained conditions, often resulting in playback stalls.

To address these challenges, we present MultiMoQ, a multi-access tile streaming framework built on MoQ that redesigns its delivery mechanism to enable robust high-bitrate streaming across multiple access paths, while supporting flexible tile scheduling and seamless access switching without playback stalls.

We implement a fully functional prototype of MultiMoQ and evaluate it in network emulation under heterogeneous real-world network conditions, comparing against Dynamic Adaptive Streaming over HTTP (DASH) and standard MoQ. Results show that MultiMoQ significantly improves robustness, increasing average goodput by about 61% and reducing tail end-to-end latency by about 69% compared to DASH, while reducing average stall time by about 97%ccompared to standard MoQ. These transport gains also translatecinto smoother viewport playback, reducing playback freeze ratio to 1.09%, compared with 3.96% for standard MoQ and 51.5% for DASH.

Reference

Yitong Li, Xinjiao Li, Ruonan Chai, Dirk Kutscher; MultiMoQ: Multi-Access Media-Over-QUIC for Robust Immersive Video Streaming; accepted for publication at ACM Multimedia 2026; Rio de Janeiro, Brazil — 10–14 November 2026

Our paper on INDS: Incremental Named Data Streaming for Real-Time Point Cloud Video has been accepted at ACM Multimedia 2025.

Abstract:

Real-time streaming of point cloud video – characterized by high data volumes and extreme sensitivity to packet loss – presents significant challenges under dynamic network conditions. Traditional connection-oriented protocols such as TCP/IP incur substantial retransmission overhead and head-of-line blocking under lossy conditions, while reactive adaptation approaches such as DASH lead to frequent quality fluctuations and a suboptimal user experience. In this paper, we introduce INDS (Incremental Named Data Streaming), a novel adaptive transmission framework that exploits the inherent layered encoding and hierarchical object structure of point cloud data to enable clients to selectively request enhancement layers based on available bandwidth and decoding capabilities. Built on Information-Centric Networking (ICN) principles, INDS employs a hierarchical naming scheme organized by time windows and Groups of Frames (GoF), which enhances cache reuse and facilitates efficient data sharing, ultimately reducing both network and server load. We implemented a fully functional prototype and evaluated it using emulated network scenarios. The experimental results demonstrate that INDS reduces end-to-end delay by up to 80%, boosts effective throughput by 15%–50% across diverse operating conditions, and increases cache hit rates by 20%–30% on average.

References

Ruonan Chai, Yixiang Zhu, Xinjiao Li, Jiawei Li, Zili Meng, Dirk Kutscher; INDS: Incremental Named Data Streaming for Real-Time Point Cloud Video; accepted for publication at ACM Multimedia 2025; October 2025

Generative AI systems are approaching a scalability limit in their development. Due to power density issues, it will soon become infeasible to train large language models with an increasing number of parameters in a single datacenter. While the industry is actively pursuing an effort to scale up AI systems, it becomes necessary to explore the use of scaled-out, global distributed systems to train or serve generative AI models.

Besides, services based on generative AI ask for stringent quality of service levels to meet users demand. Meeting those requirements can be addressed by using systems mixing powerful computing instances residing in cloud platforms with localized edge platforms, using heterogeneous and distributed systems.

Those questions may find a solution in approaches adopted by federated learning systems, in which models are trained among several stakeholders. Yet, those systems also face scalability issues in dealing with models of a larger size.

The ACM CoNEXT INet4AI workshop aims at discussing the networking challenges raised by the distribution of generative AI workloads at a large scale. To that extend, we aim at receiving contributions from academic researchers, machine learning system developers or AI infrastructure providers. .

Submitted papers must be at most six (6) pages long, excluding references and appendices, in two-column 10pt ACM format. Authors of accepted submissions are expected to present and discuss their work at the workshop. All submissions will be peer-reviewed, and the review process will be double-blind. Per the anonymity guidelines, please prepare your paper in a way that preserves the anonymity of the authors. No information will be shared with third parties.

Please submit your paper using the INET4AI Submission Portal: https://inet4ai25.hotcrp.com.

Our paper on AdaptQNet: Optimizing Quantized DNN on Microcontrollers via Adaptive Heterogeneous Processing Unit Utilization has been accepted at ACM MobiCom-2025.

Abstract

There is a growing trend in deploying DNNs on tiny micro-controller (MCUs) to provide inference capabilities in the IoT. While prior research has explored many lightweight techniques to compress DNN models, achieving overall efficiency in model inference requires not only model optimization but also careful system resource utilization for execution. Existing studies primarily leverage arithmetic logic units (ALUs) for integer-only computations on a single CPU core. Floating-point units (FPU) and multi-core capabilities available in many existing MCUs remain underutilized.

To fill this gap, we propose AdaptQNet, a novel MCU neural network system that can determine the optimal precision assignment for different layers of a DNN model. AdaptQNet models the latency of various operators in DNN models across different precisions on heterogeneous processing units. This facilitates the discovery of models that utilize FPU and multi-core capabilities to enhance capacity while adhering to stringent memory constraints. Our implementation and experiments demonstrate that AdaptQNet enables the deployment of models with better accuracy-efficiency trade-off on MCUs.

References

Yansong Sun, Jialuo He, Dirk Kutscher, Huangxun CHEN; AdaptQNet: Optimizing Quantized DNN on Microcontrollers via Adaptive Heterogeneous Processing Unit Utilization; The 31st Annual International Conference On Mobile Computing And Networking (MobiCom 2025)

Our paper on Rethinking Dynamic Networks and Heterogeneous Computing with Automatic Parallelization has been accepted by the 9th Asia-Pacific Workshop on Networking (APNET'25).

Abstract:
Hybrid parallelism techniques are crucial for the efficient training of large language models (LLMs). However, these techniques often introduce differentiated computational and communication tasks across nodes. Existing automatic parallel planning frameworks typically fail to consider both node heterogeneity and dynamic changes in network topology simultaneously, limiting their practical performance. In this paper, we address this issue by positioning heterogeneous nodes within dynamic network environments and employing a simulator to identify optimal parallel strategies. Our approach achieves fine-grained workload distribution in scenarios featuring node heterogeneity and complex networks, while also matching state-of-the-art performance in regular topologies and stable network conditions. Moreover, to mitigate the excessively long search times caused by large search spaces in existing frameworks, we propose a strategy pruning technique to rapidly eliminate infeasible parallel configurations. We further accelerate the search process by executing search tasks in parallel within the simulator. Preliminary evaluation results demonstrate that our method significantly improves training performance on heterogeneous nodes, and the proposed dynamic network design offers enhanced adaptability for complex scenarios such as cloud computing environments.

References

Ruilong Wu, Xinjiao Li, Yisu Wang, Xinyu Chen, Dirk Kutscher; Rethinking Dynamic Networks and Heterogeneous Computing with Automatic Parallelization; The 9th Asia-Pacific Workshop on Networking (APNET'25); August 2025; doi/10.1145/3735358.3735382

Our paper on ViFusion: In-Network Tensor Fusion for Scalable Video Feature Indexing has been accepted at the ACM International Conference on Multimedia Retrieval 2025 (CCF-B).

Abstract:
Large-scale video feature indexing in datacenters is critically dependent on efficient data transfer. Although in-network computation has emerged as a compelling strategy for accelerating feature extraction and reducing overhead in distributed multimedia systems, harnessing advanced networking resources at both the switch and host levels remains a formidable challenge. These difficulties are compounded by heterogeneous hardware, diverse application requirements, and complex multipath topologies. Existing methods focus primarily on optimizing inference for large neural network models using specialized collective communication libraries, which often face performance degradation in network congestion scenarios.

To overcome these limitations, we present ViFusion, a communication aware tensor fusion framework that streamlines distributed video indexing by merging numerous small feature tensors into consolidated and more manageable units. By integrating an in-network computation module and a dedicated tensor fusion mechanism within datacenter environments, ViFusion substantially improves the efficiency of video feature indexing workflows. The deployment results show that ViFusion improves the throughput of the video retrieval system by 8–22x with the same level of latency as state-of-the-art systems.

Stay tuned for the pre-print.

References

Yisu Wang, Yixiang Zhu, Dirk Kutscher; ViFusion: In-Network Tensor Fusion for Scalable Video Feature Indexing; The 15th ACM International Conference on Multimedia Retrieval; June 2025; Preprint

On Monday, December 9th, 2024, we held our Decentralization of the Internet (DIN) workshop at ACM CoNEXT-2024. It brought together network researchers, law and policy experts, and digital right activists to discuss the observed consolidation and centralization of the existing Internet applications, services, and the infrastructure in recent years. This trend has economic as well as technical implications for attributes commonly associated with the Internet, such as user-centricity and permissionless innovations.

The Decentralization of the Internet Research Group (DINRG) of the Internet Research Task Force (IRTF) has been working on identifying the root causes and consequences of Internet centralization at IETF meetings and focused workshops in the past, which has led to significant insights, especially with regard to the centralization of infrastructure and control power. This recent DIN workshop at ACM CoNEXT-2024, organized by my DINRG co-chairs Lixia Zhang and myself, provided a forum for academic researchers to present and discuss on-going efforts on this topic, and to create a greater awareness of this important issue in the broader network research community. The workshop attracted a diverse set of researchers who are working on Internet decentralization in fields such as Internet technologies, economics and law-making. The workshop featured two keynotes, two technical paper presentation sessions, and an interactive panel discussion.

Keynotes

The keynotes were presented by two renowned experts:

Keynote: Cory Doctorow: DISENSHITTIFY OR DIE!

Cory Doctorow, member of the Electronic Frontier Foundation (EFF), gave a talk titled DISENSHITTIFY OR DIE! How computer scientists can halt enshittification to make a new, good internet and condemn today's enshitternet to the scrapheap of history. Cory’s talk vividly explained the historic development of a process that he called enshittification, a process in which the providers of online products and services changed their policies subtly and gradually over time, grabbing the control of user data for profitability. Doctorow also discussed potential remedies and countermeasures, including removing the barriers for users to exit platforms and reinstalling the end-to-end principle in future application developments.

Keynote: Michael Karanicolas: The Fediverse Papers: Constitutional, Governance, and Policy Questions for a New Paradigm of Networking

Michael Karanicolas, the executive director of the UCLA Institute for Technology, Law and Policy, talked about the Fediverse Papers: Constitutional, Governance, and Policy Questions for a New Paradigm of Networking. Michael provided an overview of the history of digital speech and content governance. He highlighted the challenges in supporting effective content moderation in today’s Internet contexts, including issues around monetization, legislation, privacy, and the need for governance mechanisms to meet users, content owners, and governments’ expectations. He emphasized the importance of intentionality and a structured process to identify the essential policy questions and to evaluate various design choices for the future of decentralized platforms.

Decentralized Systems

Bluesky and the AT Protocol: Usable Decentralized Social Media

Authors: Martin Kleppmann, Paul Frazee, Jake Gold, Jay Graber, Daniel Holmgren, Devin Ivy, Jeromy Johnson, Bryan Newbold, Jaz Volpert

Abstract: Bluesky is a new social network built upon the AT Protocol, a decentralized foundation for public social media. It was launched in private beta in February 2023, and has grown to over 10 million registered users by October 2024. In this paper we introduce the architecture of Bluesky and the AT Protocol, and explain how the technical design of Bluesky is informed by our goals: to enable decentralization by having multiple interoperable providers for every part of the system; to make it easy for users to switch providers; to give users agency over the content they see; and to provide a simple user experience that does not burden users with complexity arising from the system’s decentralized nature. The system’s openness allows anybody to contribute to content moderation and community management, and we invite the research community to use Bluesky as a dataset and testing ground for new approaches in social media moderation.

ReP2P Matrix: Decentralized Relays to Improve Reliability and Performance of Peer-to-Peer Matrix

Authors: Benjamin Schichtholz, Roland Bless, Florian Jacob, Hannes Hartenstein, Martina Zitterbart

Abstract: Matrix is a decentralized middleware for low-latency group communication, most renowned for its use in the Element instant messenger. Proposals for peer-to-peer (P2P) Matrix architectures aim to decentralize the current architecture further, which is based on federated servers. These proposals require that the receiver and the originator, or another peer that already successfully received the message, are simultaneously online. We introduce relay-enhanced P2P Matrix (ReP2P Matrix) in order to improve message delivery between peers that are online at different times. The design maintains the advantages of P2P Matrix and integrates well into it, e.g., it reuses existing mechanisms for authentication and authorization. Using an extended real-world group messaging traffic dataset, we evaluate P2P Matrix by comparing it to P2P Matrix without relays. The results show that relays do not only improve reliability in message delivery, but also increase the share of low delivery latencies by 50% points in groups with up to 30 members.

On Empowering End Users in Future Networking

Authors: Tianyuan Yu, Xinyu Ma, Lixia Zhang

Abstract: In today's Internet, end users communicate largely via cloud-based apps, and user data are stored in cloud servers and controlled by cloud providers. Recent years have witnessed multiple efforts in developing decentralized social apps with various design approaches, although the community at large is yet to fully understand the effectiveness, viability, and limitations of these different designs. In this paper, we make a proposition that a necessary condition of moving towards Internet decentralization is enabling direct user-to-user (U2U) communications, and discuss the design choices in several decentralization efforts and identify their limitations. We then articulate why a DNS-derived namespace is the best choice in U2U app developments in general, and use a recently developed decentralized app, NDN Workspace (NWS), as an example to show how NWS' use of DNS-derived namespace enables secure U2U communications.

Technologies for Decentralization

Atomicity and Abstraction for Multi-Blockchain Interactions

Authors: Huaixi Lu, Akshay Jajoo, Kedar S. Namjoshi

Abstract: A blockchain enables secure, atomic transactions among untrusted parties. Atomicity is not guaranteed, however, for transactions whose operations span several blockchains; multi-chain atomicity must be enforced by a protocol. Such protocols are known only for special cases, such as cryptocurrency swaps, which are limited only to two chains. We propose a novel two-phase protocol that facilitates atomic executions of general multi-chain (>= 2) transactions. We formally analyze the protocol correctness and show that the proposed abstraction considerably simplifies the development of multi-chain applications. Our experiments with a prototype implementation show that the performance of the general atomicity protocol is comparable to that of custom-built implementations.

Communication Cost for Permissionless Distributed Consensus at Internet Scale

Authors: David Guzman, Dirk Trossen, Jörg Ott

Abstract: The diffusion of information that evolves a distributed computing state is a fundamental operation of a permissionless distributed consensus system (DCS). This permissionless participation decentralized the consensus over the distributed computing state, e.g., in cryptocurrencies and voting systems. For this, a permissionless DCS implements protocols to establish relationships among peers, which is then used to diffuse information. The relation establishment constitutes the control plane of the DCS, while the state diffusion is the data plane. The prevalent mechanism to realize both is a randomized peer-centric iterative diffusion. In this paper, we contrast this approach against a multicast-based design, focusing our comparison on the costs (bytes transmitted) for maintaining the relations, the control plane. We develop suitable models to account for those costs, parameterized through Internet-scale experimental insights we derived from existing DCS deployments. Our results show that the communication costs can be reduced by 30 times.

Towards a Decentralized Internet Namespace

Authors: Yekta Kocaogullar, Eric Osterweil, Lixia Zhang

Abstract: The Domain Name System (DNS) has been providing a decentralized global namespace to support all Internet applications and usages over the last few decades. In the recent years, a number of blockchain-based name systems have emerged with the claim of providing better namespace decentralization than DNS. The community at large seems uncertain with regard to which of these systems is the best in providing decentralized Internet namespace control. In this paper, we first deconstruct the design of DNS, identify its three essential components and explain who controls each of them. We then examine the Ethereum Name Service (ENS) as a representative example of blockchain-based naming systems, gauge the degree of its decentralization. Finally, we conduct a comparative analysis between DNS and ENS to assess the validity and affordability of each design and the (de)centralization in their namespace control and name system operations.

Panel Discussion: Decentralization of the Internet – Quo Vadis?

An interactive panel discussion with (from left to right) Michael Karanicolas (UCLA), Paul Mockapetris (ThreatSTOP), Dan Massey (USC ISI, NSF), and Cory Doctorow (EFF), articulated various next steps for countering Internet centralization. Among many things discussed, the panel and audience identified the notion of enabling direct user-to-user communication without reliance on third parties, and the required functionality to support that, such as how to provide user owned identities, tools for user mutual authentications and secure communications.

These and additional related topics will be further discussed at the IRTF DIN research group, which is a forum with open participation to serve the purpose of continuous international collaborative research on Internet decentralization.

References

• Workshop homepage at ACM CoNEXT-2024
• Workshop papers
• Workshop videos
• IETF blog post on the workshop
• IRTF DINRG

Our ACM CoNEXT-2024 workshop on the decentralization of the Internet on Monday, December 9th 2024 in LA has an exciting agenda – don't miss it! Check out the workshop homepage for up-to-date information.

09:00 Session 1: Keynotes

1. Keynote by Cory Doctorow: DISENSHITTIFY OR DIE! How computer scientists can halt enshittification to make a new, good internet and condemn today's enshitternet to the scrapheap of history.
2. Keynote by Michael Karanicolas: The Fediverse Papers: Constitutional, Governance, and Policy Questions for a New Paradigm of Networking

11:00 Session 2: Decentralized Systems

1. Martin Kleppmann, et al.; Bluesky and the AT Protocol: Usable Decentralized Social Media
2. Benjamin Schichtholz et al.; ReP2P Matrix: Decentralized Relays to Improve Reliability and Performance of Peer-to-Peer Matrix
3. Tianyuan Yu et al.; On Empowering End Users in Future Networking

14:00 Session 3: Technologies for Decentralization

1. Huaixi Lu et al.; Atomicity and Abstraction for Multi-Blockchain Interactions
2. David Guzman et. el; Communication Cost for Permissionless Distributed Consensus at Internet Scale
3. Yekta Kocaogullar et al.; Towards a Decentralized Internet Namespace

15:00 Session 4: Decentralization of the Internet – Quo Vadis?

• Organizers: Lixia Zhang & Dirk Kutscher
• Interactive panel discussion with Cory Doctorow, Michael Karanicola, and paper authors

The ACM Information-Centric Networking 2022 Conference took place in Osaka from September 19 to 21 2022, hosted by Osaka University. It was a three-day conference with tutorials, one keynote, two panel session, and paper and poster/demo presentations. The highlights (with links to papers and presentations) from my perspective were the following:

Keynote by Dave Oran: Travels with ICN – The road traversed and the road ahead

Dave Oran presented an overview of his research experience over the last ten years that was informed by many seminal research contributions on ICN and his career in the network vendor sector as well as in standards and research bodies such as the IETF and IRTF.

The keynote's theme was about disentagling the application and network layer aspects of ICN, which led to interesting perspectives on some of the previous design decisions in CCNx and NDN.

As ilustrated in the figure below, the more networking-minded ICN topics are typically connected to features and challenges of building packet-forwarding networks based on the principle of accessing named data. The actual research questions are generally not different to those of IP networks (routing, mobility etc.), but ICN provides a significant potential to re-think and often improve over the specific approaches in IP networks due to its core properties such as object security and symmetric, stateful forwarding.

Information-centric applications development in contrast is often concerned with general naming concepts, namespace design, and security features that are enabled by namespace design and application layer object security such as trust schema and provenance.

The message in Dave's talk was not that these are completely disjunct areas that should best be investigated independent of each other, but rather that the ICN's fascination and disruptive potential is based on the potential for rethinking layer boundaries and contemplating a better function split between applications, network stacks on endpoints, and forwarding elements in the network. In his talk, Dave focused on

• the Interaction of consumers & networking producers of data;
• routing;
• forwarding; and
• congestion control.

He discussed many lessons learned as well as open research and new ideas for all of these topics – please refer to the presentation slides for details.

One particularly interesting current ICN research topic is distributed computing and ICN architectures & interaction models for that. ICN's name-based forwarding model and object security provide very interesting options for simplifying systems such as microservices, RESTful services and distributed application coordination. Alluding to our work on Reflexive Forwarding, Dave offered two main lessons learned from building corresponding communication abstractions:

1. Content fetch with two-way handshakes is a poor match for doing distributed computations.

2. Extensions to the base protocols can give a flexible underpinning for multiple interaction models

This raises the question of the slim waist of ICN, i.e., as research progresses, what should be the minimal feature set and what is the right extensibility model?

Dave concluded his talk with a few interesting questions:

• how can the networking insights we’ve gained from ICN protocols inform the construction of Information Centric systems and applications?

  • Whether and how to utilize name-based routing to achieve robustness and performance scaling for distributed applications?
  • Where does caching help or not help and how to best utilize caches?
  • Does pushing Names down to lower layers help latency? Resilience? Fairness?

• How can the insights we’ve gained from applying Information Centricity in applications inform what we bother to change the network to do, and what not?

  • Do things like multipath forwarding, in-network retransmission, or reflexive forwarding actually enable applications that are hard or infeasible to do without them?
  • Is there a big win for wireless networks in terms of optimizing a scarce resource or having more robust and responsive mobility characteristics?

More details in the presentation slides

Panel: ICN and the Metaverse – Challenges and Opportunities

I had the pleasure of being in a panel with Jeff Burke (UCLA) and Geoff Houston (APNIC), moderated by Alexander Afanasyev (Florida International University) discussing Metaverse challenges and opportunities for ICN.

Questions on Metaverse and ICN

Large-scale interactive and networked AR/VR/XR systems are now referred to as Metaverse, and the general assumption is that corresponding applications will be hosted on platforms, similar to those that are employed for web and social media applications today.

In the web, the platform approach has led to an accelerated development and growth of a few popular mainstream systems. On the other hand, several problems have been observed such as ubiquitous surveillance, lock-in effects, centralization, innovation stagnation, and cost overhead for achieving the required performance.

While these phenomena may have both technical and economic root causes, we would like to discuss:

• How should Metaverse systems be designed, and what would be important architectural pillars?
• What is the potential for re-imagining Metaverse with information-centric concepts and protocols?
• Would ICN enable or lead to profound architecturally unique approaches – or would protocols such as NDN be a drop-in replacement for QUIC, HTTP3 etc.?
• What are the challenges for building ICN-based Metaverse systems, and what it missing in today's ICN platforms?

As input to the discussion, Jeff Burke and myself (together with Dave Oran) submitted two papers:

• Jeff Burke: Statement: As TCP/IP is to the web, ICN is to the…?
• Dirk Kutscher and David Oran: RESTful information-centric networking: statement

Research Directions

Jeff offered a list of really interesting research directions based on the notion that in the Metaverse, host-based identifiers and end-to-end connections between hosts would be abstracted even further away than in today’s web. Client devices would fade into the background in favor of the data supplanting or augmenting the real world. Thus, a metaverse consisted of information not associated with the physical world unless it needed to describe or provide interaction with it. The experiential semantics were viscerally information-centric, which would help to motivate the ICN research opportunities such as:

• Persistence: The information forming a metaverse persists across sessions and users.

• “Content” and Interoperability: Designing the relationships among metaverse-layer objects and the named packets that an ICN network moves and stores.

• Naming and Spatial Organization: How to best integrate knowledge from research in databases and related fields where these challenges have been considered for decades.

• Trust, Provenance, and Transactions: Using ICN to disentangle metaverse objects from the security provided by a source or a given channel of communication, with the named data representation secured at the time of publication instead.

RESTful ICN

In our paper on RESTFul ICN, Dave Oran and I asked the question: given that most web applications are concerned with transferring named units of data (web resources, video chunks etc.), can the REST paradigm be married with the data-oriented, receiver-driven operation of Information-Centric Networking (ICN), leveraging attractive ICN benefits such as consumer anonymity, stateful and symmetric forwarding, flow-balance in-network caching, and implicit object security?

We argue that this is feasible given some of the recent advances in ICN protocol development and that the resulting suite is simpler and potentially having better performance and robustness properties. Our sketch of an ICN based protocol framework addresses secure and efficient establishment and continuation of REST communication sessions, without giving up key ICN properties, such as consumer anonymity and flow balance.

Panel Discussion

The panel discussed the current socio-economic realities in the Internet and the Web and explored opportunities (and non-opportunities) for redesigns, and how ICN could be a potential enabler for that.

My personal view is that most of the potential dystopian outcomes of future Metaverse applications are independent from the enabling networking technology and the technology stack at large (security, naming etc.). It is really important to understand the actual objectives of a specific systems, i.e., who operates to which ends, similar to so-called social networks today. If the main objective is to create a more powerful advertising and manipulation platform, then such as system will exhibit yet unimaginable surveillance and tracking mechanisms – independent of the underlying network stack.

With respect to the technical design, I agree to Jeff Burke's proposed research directions. One particularly interesting question will be how to design a Information-Centric communication stack and corresponding APIs. I argued that it is not necessary to replicate existing interaction styles and protocol stacks from the TCP/IP (or QUIC) world. Instead it should be more interesting and productive to discuss the fundamentally needed interaction classes such as

• High-performance multi-destination transfer
• Group communication and synchronization
• High-performance session-oriented communication with servers and peers (for which we proposed RESTful ICN).

The panel then also discussed how likely non-mainstream Metaverse systems would be adopted and whether the current socio-economic environment actually allows for that level of permissionless innovation – considering the network effects that Metaverse systems would be subjected to, much in the same way as so-called social networks.

Panel: Hard Lessons for ICN from IP Multicast?

Thomas Schmidt (HAW Hamburg) moderated a panel discussion with Jon Crowcroft (University of Cambridge), Dave Oran, and George Xylomenos (Athens University of Economics and Business) as panelists.

With the continued shift towards more and more live video streaming services over the Internet, scalable multi-destination delivery has become more relevant again. For example, the recently chartered IETF Working Group on Media over QUIC (MOQ), is addressing the need for scalable multi-destination delivery and the unavailability of IP multicast as a platform by developing a QUIC-based overlay system that essentially uses information-centric concepts, albeit in a QUIC overlay network. Such system would consist of a network of QUIC proxies, connected via individual QUIC connections to emulate request forwarding and chunk-based video data distribution. Considering the non-negligible overhead and complexity one might ask the question whether live video streaming over the Internet could be served by a better approach. Questions like this are being asked by the network service provider community (ISPs have to bear a lot of the overhead and overlay complexity) as well, for example in this APNIC blog posting by Jake Holland titled Why inter-domain multicast now makes sense.

This panel was inspired by a statement paper submitted by Jon Crowcroft titled [Hard lessons for ICN from IP multicast (https://dl.acm.org/doi/10.1145/3517212.3558086). In this brief statement, Jon traced the line of thought from Internet multicast through to Information Centric Networking, and used this to outline what he thinks should have been the priorities in ICN work from the start.

The statement paper discusses a few problems with IP multicast that have been largely acknowledged such as difficulties in creating viable business models, unsolved security problems such as IP multicast being used as a DDOS platform, and interdomain multicast that proven difficult to establish due multicast routing scaling problems and the lack of robust pricing models. The second part of the paper is then some ICN work that has been addressing some of the mentioned issued.

The paper gave rise to an interesting and controversial discussion at the panel. The most important point is IMO to characterize ICN communication model correctly: it is correct that the combination of stateful forwarding, Interest aggregation, and caching enables an implicit multi-destination delivery service. It is implicit, because consumers that ask for the same units of named data within a time frame at the order of the network RTT will send equivalent Interest messages so that forwarders can multicast the data delivery to the faces they received such Interests from. In conjunction with opportunistic (or managed) caching by forwarders this would enable a very elegant multi-destination delivery services that can even cater to a wider variation of Interest sending times, as "late" Interest would be answered from caches.

This is a different service model compared to the push-based IP multicast model. ICN does not provide such as service in the first place, but is just applying its regular receiver-driven mode of operation which elegantly works well in the case of multiple consumers asking for the same data. It is probably fair to say that the ICN model caters to media-delivery use cases (one stream delivered to multiple consumers) but does not try to provide the more general IP multicast service model (Any Source Multicast). However, by extension, the ICN approach could be applied to multi-source scenarios as well – the system would build implicit delivery trees from any source to current consumers, without requiring extra machinery.

With this, if you like, simpler service model, ICN does fundamentally not inherit many of the problems that prohibit IP multicast in the Internet: the system is receiver-driven which simply eliminates DDOS threats (on the packet level). It is also not clear, whether ICN would need anything special to provide this service in inter-domain settings (except for general ICN routing in the Internet, which is a general,
but different research question).

Acknowledging this conceptual and practical difference, there are obviously other interesting research questions that ICN multi-destination delivery entails, for example performance and jitter reduction in the presence of caching and other transport questions.

Overall, a good time to talk about multi-destination delivery and to keep thinking about missing pieces and potential future work in ICN.

Enabling Distributed Applications

One paper presentation session was focused on distributed applications – a very interesting and relevant ICN research area. It featured three great papers:

SoK: The evolution of distributed dataset synchronization solutions in NDN

This paper by Philipp Moll, Varun Patil, Lan Wang, and Lixia Zhang systemizes the knowledge about distributed dataset synchronisation in ICN, or Sync in short, which, according to the authors, plays the role of a transport service in the Named Data Networking (NDN) architecture. A number of NDN Sync protocols have been developed over the last decade. For this paper, they conducted a systematic examination of NDN Sync protocol designs, identified common design patterns, revealed insights behind different design approaches,
and collected lessons learned over the years.

Sync enables new ways of thinking about coordination and general communication in distributed ICN systems, and I encourage everyone to read this for a good overview of the different proposed systems and their properties.

There are also some open research questions around Sync, such as large-scale applicability, alternative to using Interest multicast for discovery and more – a good topic to work on!

DICer: distributed coordination for in-network computations

This paper by Uthra Ambalavanan, Dennis Grewe, Naresh Nayak, Liming Liu, Nitinder Mohan, and Jörg Ott is a nice product of the Piccolo project that had the pleasure to set up and co-lead.

Application domains such as automotive and the Internet of Things may benefit from in-network computing to reduce the distance data travels through the network and the response time. Information Centric Networking (ICN) based compute frameworks such as Named Function Networking (NFN) are promising options due to their location independence and loosely-coupled communication model.

However, unlike current operations, such solutions may benefit from orchestration across the compute nodes to use the available resources in the network better. In this paper, the authors adopted the State Vector Synchronization (SVS), an application dataset synchronization protocol in ICN, to enhance the neighborhood knowledge of in-network compute nodes in a distributed fashion. They designed distributed coordination for in-network computation (DICer) that assists the service deployments by improving the resolution of compute requests.

Kua: a distributed object store over named data networking

This paper by Varun Patil, Hemil Desai, and Lixia Zhang decribes a distributed object store in NDN.

Applications such as machine learning training systems or log collection generate and consume large amounts of data. Object storage systems provide a simple abstraction to store and access such large datasets. These datasets are typically larger than the capacities of individual storage servers, and require fault tolerance through replication. This paper presents Kua, a distributed object storage system built over Named Data Networking (NDN).

The data-centric nature of NDN helps Kua maintain a simple design while catering to requirements of storing large objects, providing fault tolerance, low latency and strong consistency guarantees, along with data-centric security.

ICN Applications and Wireless Networking

The session on ICN Applications and Wireless Networking features four papers:

N-DISE: NDN-based data distribution for large-scale data-intensive science

This paper by Yuanhao Wu, Faruk Volkan Mutlu, et al. describes an NDN for Data-Intensive Science Experiments (N-DISE).

To meet unprecedented challenges faced by the world’s largest data- and network-intensive science programs, the authors designed and implemented a new, highly efficient and field-tested data distribution, caching, access and analysis system for the Large Hadron Collider (LHC) high energy physics (HEP) network and other major science programs. They developed a hierarchical Named Data Networking (NDN) naming scheme for HEP data, implemented new consumer and producer applications to interface with the high-performance NDNDPDK forwarder, and buildt on recently developed high-throughput NDN caching and forwarding methods.

The experiemts in this paper include delivering LHC data over the wide area network (WAN) testbed at throughputs exceeding 31 Gbps between Caltech and StarLight, with dramatically reduced download time.

Building a secure mHealth data sharing infrastructure over NDN

In this paper Saurab Dulal, Nasir Ali, et al. describes an NDN-based mHealth system called mGuard.

Exploratory efforts in mobile health (mHealth) data collection and sharing have achieved promising results. However, fine-grained contextual access control and real-time data sharing are two of the remaining challenges in enabling temporally-precise mHealth intervention. The authors have developed an NDN based system called mGuard to address these challenges. mGuard provides a pub-sub API to let users subscribe to real-time mHealth data streams, and uses name-based access control policies and key-policy attribute-based encryption to grant fine-grained data access to authorized users based on contextual information.

Delay-tolerant ICN and its application to LoRa

I have co-authored this paper together with Peter Kietzmann, José Alamos, Thomas C. Schmidt, and Matthias Wählisch.

Connecting low-power long-range wireless networks, such as LoRa, to the Internet imposes significant challenges because of the vastly longer round-trip-times (RTTs) in these constrained networks. In our paper on "Delay-Tolerant ICN and Its Application to LoRa" we present an Information-Centric Networking (ICN) protocol framework that enables robust and efficient delay-tolerant communication to edge networks, including but not limited to LoRa. Our approach provides ICN-idiomatic communication between networks with vastly different RTTs for different use cases. We applied this framework to LoRa, enabling end-to-end consumer-to-LoRa-producer interaction over an ICN-Internet and asynchronous ("push") data production in the LoRa edge. Instead of using LoRaWAN, we implemented an IEEE 802.15.4e DSME MAC layer on top of the LoRa PHY layer and ICN protocol mechanisms in the RIOT operating system.

For our experiments, we connected constrained LoRa nodes and gateways on IoT hardware platforms to a regular, emulated ICN network and performed a series of measurements that demonstrate robustness and efficiency improvements compared to standard ICN.

iCast: dynamic information-centric cross-layer multicast for wireless edge network

This paper by Tianlong Li, Tian Song, Yating Yang, and Jike Yang presents iCast, short for dynamic information-centric multicast, to enable dynamic multicast in the link layer.

Native multicast support in Named Data Networking (NDN)
is an attractive feature, as multicast content delivery can reduce the redundant traffic and improve the network performance, especially in wireless edge networks. With their visibility into Interest and Data names, NDN routers automatically aggregate the same requests from different end hosts and establish network-layer multicast. However,
the current link-layer multicast based on host-centric MAC address management is inflexible. Consequently, supporting NDN dynamic multicast with the current link-layer architecture remains a challenge.

iCast enables dynamic multicast in the link layer based on three main contributions:

1. iCast integrates NDN native multicast with the host-centric link layer while maintaining the host-centric properties of the current link layer.
2. iCast achieves per-packet dynamic multicast in the link layer, and the authors further propose a hash-based iCast variant for dynamic connection.
3. iCast has been implemented in a real testbed, and the evaluation results show that iCast reduces up to 59.53% traffic compared with vanilla NDN. iCast bridges the gap between NDN multicast and the host-centric link-layer multicast.