Nitinder Mohan
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
51 records found
1
Low Earth Orbit (LEO) satellite ISPs promise universal Internet connectivity, yet their interaction with content delivery remains poorly understood. We present the first comprehensive measurement study decomposing Starlink's web content delivery performance decomposed across Point of Presence (PoP), DNS, and CDN layers. To quantify how satellite architecture disrupts terrestrial CDN assumptions, we conduct a measurement study spanning two years. We identify three distinct performance regimes based on infrastructure density. Regions with local content-rich PoPs achieve near-terrestrial latencies with the satellite segment dominating 80-90% of RTT. Infrastructure-sparse regions suffer cascading penalties: remote PoPs force distant resolver selection, which triggers CDN mislocalization, pushing latencies beyond 200 ms. Dense-infrastructure regions show minimal sensitivity to PoP changes. Leveraging Starlink's infrastructure expansion in early 2025 as a natural experiment, we demonstrate that relocating PoPs closer to user location reduces median page-fetch times by 60%. Our findings reveal that infrastructure proximity, not satellite coverage, influences web performance, requiring fundamental changes to CDN mapping and DNS resolution for satellite ISPs.
...
Low Earth Orbit (LEO) satellite ISPs promise universal Internet connectivity, yet their interaction with content delivery remains poorly understood. We present the first comprehensive measurement study decomposing Starlink's web content delivery performance decomposed across Point of Presence (PoP), DNS, and CDN layers. To quantify how satellite architecture disrupts terrestrial CDN assumptions, we conduct a measurement study spanning two years. We identify three distinct performance regimes based on infrastructure density. Regions with local content-rich PoPs achieve near-terrestrial latencies with the satellite segment dominating 80-90% of RTT. Infrastructure-sparse regions suffer cascading penalties: remote PoPs force distant resolver selection, which triggers CDN mislocalization, pushing latencies beyond 200 ms. Dense-infrastructure regions show minimal sensitivity to PoP changes. Leveraging Starlink's infrastructure expansion in early 2025 as a natural experiment, we demonstrate that relocating PoPs closer to user location reduces median page-fetch times by 60%. Our findings reveal that infrastructure proximity, not satellite coverage, influences web performance, requiring fundamental changes to CDN mapping and DNS resolution for satellite ISPs.
PRISM
Cross-Layer Observability Framework for Mobile Core Networks
Cloud-native 5G core networks span service, orchestration, and infrastructure layers, each producing telemetry in different formats, at different rates, with different semantics. Yet monitoring tools operate per layer, leaving operators without cross-layer visibility needed to diagnose faults that cascade across architectural boundaries. We present PRISM, the first-of-its-kind cross-layer observability framework designed to plug into existing cloud-native cellular core network deployments, capturing evolving structural and temporal dependencies. PRISM collects and correlates logs, metrics, and configuration state from three layers into a unified, continuously updated representation, using an ontology-driven temporal knowledge graph (TKG) as core data structure. To support fault investigation, PRISM provides an incident-time subgraph extraction algorithm that, given incident time and rollback window, produces a compact diagnostic view of the TKG by highlighting anomalous entities and dependency paths connecting them. We evaluate PRISM on an Open5GS 5G core deployment under four fault scenarios spanning resource stress and configuration changes. Across all scenarios, PRISM produces operator-usable incident-time subgraphs that capture dominant anomalous components and their dependency structure, enabling focused troubleshooting and accelerating issue resolution time in virtualized mobile core networks.
...
Cloud-native 5G core networks span service, orchestration, and infrastructure layers, each producing telemetry in different formats, at different rates, with different semantics. Yet monitoring tools operate per layer, leaving operators without cross-layer visibility needed to diagnose faults that cascade across architectural boundaries. We present PRISM, the first-of-its-kind cross-layer observability framework designed to plug into existing cloud-native cellular core network deployments, capturing evolving structural and temporal dependencies. PRISM collects and correlates logs, metrics, and configuration state from three layers into a unified, continuously updated representation, using an ontology-driven temporal knowledge graph (TKG) as core data structure. To support fault investigation, PRISM provides an incident-time subgraph extraction algorithm that, given incident time and rollback window, produces a compact diagnostic view of the TKG by highlighting anomalous entities and dependency paths connecting them. We evaluate PRISM on an Open5GS 5G core deployment under four fault scenarios spanning resource stress and configuration changes. Across all scenarios, PRISM produces operator-usable incident-time subgraphs that capture dominant anomalous components and their dependency structure, enabling focused troubleshooting and accelerating issue resolution time in virtualized mobile core networks.
ARTA
Adaptive Redundancy-aware Telemetry node Activation in 6G Edge
Future 6G edge networks must support ultra-reliable and low-latency communication for applications such as autonomous driving and industrial automation, while operating under strict bandwidth and compute constraints. Achieving this balance requires telemetry systems that are adaptive, scalable, and resource-efficient. However, existing telemetry approaches either rely on static deployment or network-wide activation, leading to excessive overhead and redundant measurements in dense edge environments. This paper presents ARTA, an Adaptive Redundancy-Aware Telemetry Node Activation framework that formulates the telemetry activation problem as a redundancy-aware benefit-cost optimization. Each node is evaluated by anomaly likelihood, traffic importance, topological centrality, and context relevance, while redundancy, caused by nearby nodes reporting overlapping telemetry data, and resource limits are explicitly modeled. The resulting submodular objective enables a greedy algorithm with provable optimality guarantees. Simulations on edge-core topologies of up to 1,000 nodes show that ARTA achieves approximately 0.8 normalized accuracy (80% of the maximum achievable accuracy) while activating only 20% of nodes within a few seconds, demonstrating scalability and zero-touch observability.
...
Future 6G edge networks must support ultra-reliable and low-latency communication for applications such as autonomous driving and industrial automation, while operating under strict bandwidth and compute constraints. Achieving this balance requires telemetry systems that are adaptive, scalable, and resource-efficient. However, existing telemetry approaches either rely on static deployment or network-wide activation, leading to excessive overhead and redundant measurements in dense edge environments. This paper presents ARTA, an Adaptive Redundancy-Aware Telemetry Node Activation framework that formulates the telemetry activation problem as a redundancy-aware benefit-cost optimization. Each node is evaluated by anomaly likelihood, traffic importance, topological centrality, and context relevance, while redundancy, caused by nearby nodes reporting overlapping telemetry data, and resource limits are explicitly modeled. The resulting submodular objective enables a greedy algorithm with provable optimality guarantees. Simulations on edge-core topologies of up to 1,000 nodes show that ARTA achieves approximately 0.8 normalized accuracy (80% of the maximum achievable accuracy) while activating only 20% of nodes within a few seconds, demonstrating scalability and zero-touch observability.
SMOOTH
Scalable Multitask Offloading with Backbone Sharing
Intelligent mobile applications are often constrained by limited on-device hardware and by the latency and bandwidth overhead of cloud offloading. Offloading computation-intensive deep learning tasks to edge servers can potentially mitigate these challenges. However, existing systems struggle to scale as the numbers of tasks and users grow. This limitation stems from a fundamental conflict: requests in edge settings are typically sparse, heterogeneous, and demand immediate processing to minimize latency. However, GPUs operate most efficiently on dense, homogeneous batches, a condition that edge traffic rarely provides. Most existing approaches, including cloud-oriented solutions like Multi-Instance GPU, fail to resolve this tension because edge devices typically lack the advanced features available in high-end cloud GPUs. This paper contributes SMOOTH, a scalable multitask offloading system that introduces a Sparsity-to-Density Abstraction to resolve this conflict. By decomposing deep learning models into a shared backbone and lightweight task-specific heads, SMOOTH transforms sparse, heterogeneous request streams into dense, homogeneous computation blocks amenable to efficient batching. This allows highly efficient cross-task batching without prohibitive accumulation delays. To further optimize responsiveness, SMOOTH employs a compute-lightweight (O(1)), sparsity-aware adaptive scheduler that dynamically balances inference throughput and end-to-end latency based on queue dynamics. Our evaluations demonstrate up to 2.21× higher throughput, 45% lower memory usage, and up to 82% reduced latency across diverse arrival patterns compared to baselines.
...
Intelligent mobile applications are often constrained by limited on-device hardware and by the latency and bandwidth overhead of cloud offloading. Offloading computation-intensive deep learning tasks to edge servers can potentially mitigate these challenges. However, existing systems struggle to scale as the numbers of tasks and users grow. This limitation stems from a fundamental conflict: requests in edge settings are typically sparse, heterogeneous, and demand immediate processing to minimize latency. However, GPUs operate most efficiently on dense, homogeneous batches, a condition that edge traffic rarely provides. Most existing approaches, including cloud-oriented solutions like Multi-Instance GPU, fail to resolve this tension because edge devices typically lack the advanced features available in high-end cloud GPUs. This paper contributes SMOOTH, a scalable multitask offloading system that introduces a Sparsity-to-Density Abstraction to resolve this conflict. By decomposing deep learning models into a shared backbone and lightweight task-specific heads, SMOOTH transforms sparse, heterogeneous request streams into dense, homogeneous computation blocks amenable to efficient batching. This allows highly efficient cross-task batching without prohibitive accumulation delays. To further optimize responsiveness, SMOOTH employs a compute-lightweight (O(1)), sparsity-aware adaptive scheduler that dynamically balances inference throughput and end-to-end latency based on queue dynamics. Our evaluations demonstrate up to 2.21× higher throughput, 45% lower memory usage, and up to 82% reduced latency across diverse arrival patterns compared to baselines.
Horizon
Understanding and Predicting Global Starlink Performance
Starlink has deployed over 7,800 satellites serving millions of subscribers, yet predicting its performance remains an open challenge. Rapid orbital dynamics, frequent handovers, and weather-induced signal attenuation create variability that existing models, built on a handful of instrumented terminals in limited regions, cannot capture at global scale. We present Horizon, the first global-scale machine learning system for predicting LEO satellite Internet performance. Our key insight is that crowdsourced measurement platforms, while noisier than controlled experiments, provide the geographic diversity necessary to build globally generalizable models. Horizon integrates 11 months of measurements from M-Lab and Cloudflare spanning 90+ countries with meteorological data and satellite orbital propagation features. On a fully held-out one-week temporal window, Horizon achieves mean absolute errors of 17.76 ms for latency and 25.63 Mbps for throughput; on a standard 80/20 split it outperforms all baselines, including adaptations of state-of-the-art architectures. Feature importance analysis reveals that geographic position dominates prediction, with latitude alone contributing 42-46%, while weather features account for 14-15%, quantifying the impact of atmospheric conditions on Ku/Ka-band links. Leave-one-location-out experiments confirm that Horizon generalizes to regions absent from training, enabling performance estimation where measurement infrastructure does not yet exist. Our dataset and pipeline are publicly available, providing a foundation for global LEO network performance visibility.
...
Starlink has deployed over 7,800 satellites serving millions of subscribers, yet predicting its performance remains an open challenge. Rapid orbital dynamics, frequent handovers, and weather-induced signal attenuation create variability that existing models, built on a handful of instrumented terminals in limited regions, cannot capture at global scale. We present Horizon, the first global-scale machine learning system for predicting LEO satellite Internet performance. Our key insight is that crowdsourced measurement platforms, while noisier than controlled experiments, provide the geographic diversity necessary to build globally generalizable models. Horizon integrates 11 months of measurements from M-Lab and Cloudflare spanning 90+ countries with meteorological data and satellite orbital propagation features. On a fully held-out one-week temporal window, Horizon achieves mean absolute errors of 17.76 ms for latency and 25.63 Mbps for throughput; on a standard 80/20 split it outperforms all baselines, including adaptations of state-of-the-art architectures. Feature importance analysis reveals that geographic position dominates prediction, with latitude alone contributing 42-46%, while weather features account for 14-15%, quantifying the impact of atmospheric conditions on Ku/Ka-band links. Leave-one-location-out experiments confirm that Horizon generalizes to regions absent from training, enabling performance estimation where measurement infrastructure does not yet exist. Our dataset and pipeline are publicly available, providing a foundation for global LEO network performance visibility.
Networking is critical to a holistic metaverse system given the high-throughput and low-latency requirements and often distributed nature of metaverses. This chapter discusses several important aspects and directions that can enhance the networking performance in the metaverse, including user experienced delay, edge computing, multimodal networking, semantic and goal/deadline-aware networking, multipath networking, CDN, etc.
...
Networking is critical to a holistic metaverse system given the high-throughput and low-latency requirements and often distributed nature of metaverses. This chapter discusses several important aspects and directions that can enhance the networking performance in the metaverse, including user experienced delay, edge computing, multimodal networking, semantic and goal/deadline-aware networking, multipath networking, CDN, etc.
Tokens, Not Packets
Rethinking the Multipath QUIC Scheduling Interface
The Multipath QUIC extension has enabled a growing class of stream-aware schedulers that exploit QUIC's multiplexed streams, mixed reliable and unreliable delivery, and explicit control frames. Progress on these ideas is bottlenecked by an interface inherited from Multipath TCP, in which the scheduler is a function called during packet generation to pick a path. This coupling yields invasive, library-specific scheduler implementations, hinders application-steered scheduling, and makes fair cross-scheduler comparison impractical. We present an up-front, token-based MPQUIC scheduling interface designed primarily for research use, guided by three design choices: unified up-front decisions that address path selection, stream assignment, and duplication jointly; inspectability of scheduling policy as persistent data structures; and mechanism safety through a boundary that preserves protocol invariants. The scheduler reads transport state through a mediating ConnState object and writes policy into per-path, multi-level queues of abstract tokens that the library drains into packets. We prototype the interface in Cloudflare's quiche, reproduce published scheduler behavior, and demonstrate a new degree of freedom - explicit control frame routing - that reduces median stream completion time by 138 ms in our evaluation.
...
The Multipath QUIC extension has enabled a growing class of stream-aware schedulers that exploit QUIC's multiplexed streams, mixed reliable and unreliable delivery, and explicit control frames. Progress on these ideas is bottlenecked by an interface inherited from Multipath TCP, in which the scheduler is a function called during packet generation to pick a path. This coupling yields invasive, library-specific scheduler implementations, hinders application-steered scheduling, and makes fair cross-scheduler comparison impractical. We present an up-front, token-based MPQUIC scheduling interface designed primarily for research use, guided by three design choices: unified up-front decisions that address path selection, stream assignment, and duplication jointly; inspectability of scheduling policy as persistent data structures; and mechanism safety through a boundary that preserves protocol invariants. The scheduler reads transport state through a mediating ConnState object and writes policy into per-path, multi-level queues of abstract tokens that the library drains into packets. We prototype the interface in Cloudflare's quiche, reproduce published scheduler behavior, and demonstrate a new degree of freedom - explicit control frame routing - that reduces median stream completion time by 138 ms in our evaluation.
Horizon
Understanding and Predicting Global Starlink Performance
Starlink has deployed over 7,800 satellites serving millions of subscribers, yet predicting its performance remains an open challenge. Rapid orbital dynamics, frequent handovers, and weather-induced signal attenuation create variability that existing models, built on a handful of instrumented terminals in limited regions, cannot capture at global scale. We present Horizon, the first global-scale machine learning system for predicting LEO satellite Internet performance. Our key insight is that crowdsourced measurement platforms, while noisier than controlled experiments, provide the geographic diversity necessary to build globally generalizable models. Horizon integrates 11 months of measurements from M-Lab and Cloudflare spanning 90+ countries with meteorological data and satellite orbital propagation features. On a fully held-out one-week temporal window, Horizon achieves mean absolute errors of 17.76 ms for latency and 25.63 Mbps for throughput; on a standard 80/20 split it outperforms all baselines, including adaptations of state-of-the-art architectures. Feature importance analysis reveals that geographic position dominates prediction, with latitude alone contributing 42-46%, while weather features account for 14-15%, quantifying the impact of atmospheric conditions on Ku/Ka-band links. Leave-one-location-out experiments confirm that Horizon generalizes to regions absent from training, enabling performance estimation where measurement infrastructure does not yet exist. Our dataset and pipeline are publicly available, providing a foundation for global LEO network performance visibility.
...
Starlink has deployed over 7,800 satellites serving millions of subscribers, yet predicting its performance remains an open challenge. Rapid orbital dynamics, frequent handovers, and weather-induced signal attenuation create variability that existing models, built on a handful of instrumented terminals in limited regions, cannot capture at global scale. We present Horizon, the first global-scale machine learning system for predicting LEO satellite Internet performance. Our key insight is that crowdsourced measurement platforms, while noisier than controlled experiments, provide the geographic diversity necessary to build globally generalizable models. Horizon integrates 11 months of measurements from M-Lab and Cloudflare spanning 90+ countries with meteorological data and satellite orbital propagation features. On a fully held-out one-week temporal window, Horizon achieves mean absolute errors of 17.76 ms for latency and 25.63 Mbps for throughput; on a standard 80/20 split it outperforms all baselines, including adaptations of state-of-the-art architectures. Feature importance analysis reveals that geographic position dominates prediction, with latitude alone contributing 42-46%, while weather features account for 14-15%, quantifying the impact of atmospheric conditions on Ku/Ka-band links. Leave-one-location-out experiments confirm that Horizon generalizes to regions absent from training, enabling performance estimation where measurement infrastructure does not yet exist. Our dataset and pipeline are publicly available, providing a foundation for global LEO network performance visibility.
Microservice architectures allow developers to decompose their applications into independently deployable functional blocks, each with its own requirements. In order to support a wide range of constraints, service virtualization can be customized across microservices but is typically homogeneous within a cluster. As there is no clear one size fit all approach, we can improve resource utilization and performance by using virtualization as a new dimension in orchestration, especially in edge computing environments. For instance, Unikernels represent a lightweight virtualization technology that offers a performant alternative to traditional containers. While we find different studies analyzing and comparing these virtualization technologies, (a) the performance results might vary when including the overhead of the orchestration platform, and (b) it's not trivial to select the perfect virtualization technology for an entire cluster. In this paper, we explore the benefits of hybrid container-unikernel deployments by extending an orchestration framework for edge computing to allow for seamless mixing and matching of both technologies. Our evaluation shows how hybrid deployments can lead up to 44% CPU reduction cluster-wide while there are scenarios where containers are still preferable.
...
Microservice architectures allow developers to decompose their applications into independently deployable functional blocks, each with its own requirements. In order to support a wide range of constraints, service virtualization can be customized across microservices but is typically homogeneous within a cluster. As there is no clear one size fit all approach, we can improve resource utilization and performance by using virtualization as a new dimension in orchestration, especially in edge computing environments. For instance, Unikernels represent a lightweight virtualization technology that offers a performant alternative to traditional containers. While we find different studies analyzing and comparing these virtualization technologies, (a) the performance results might vary when including the overhead of the orchestration platform, and (b) it's not trivial to select the perfect virtualization technology for an entire cluster. In this paper, we explore the benefits of hybrid container-unikernel deployments by extending an orchestration framework for edge computing to allow for seamless mixing and matching of both technologies. Our evaluation shows how hybrid deployments can lead up to 44% CPU reduction cluster-wide while there are scenarios where containers are still preferable.
Emerging Low Earth Orbit (LEO) satellite constellations have been considered for uses beyond plain Internet access, including content caching and edge computing. Assuming satellites are equipped with inter-satellite links, we propose using these links and thus the space in-between satellites, paired with a dedicated satellite queuing system, to "store"data and provide access by keeping data in constant flux around the globe. We describe the properties and explore the capabilities of such a system and discuss some potential uses.
...
Emerging Low Earth Orbit (LEO) satellite constellations have been considered for uses beyond plain Internet access, including content caching and edge computing. Assuming satellites are equipped with inter-satellite links, we propose using these links and thus the space in-between satellites, paired with a dedicated satellite queuing system, to "store"data and provide access by keeping data in constant flux around the globe. We describe the properties and explore the capabilities of such a system and discuss some potential uses.
Conference paper
(2025)
-
Wei Geng, Oguz Kagan Altas, David Guzman, Giovanni Bartolomeo, Nitinder Mohan, Joerg Ott
Edge computing orchestration faces significant challenges due to resource constraints, highly distributed topologies, and dynamic network conditions. The discrepancy between theoretical and actual runtime performance often leads to suboptimal deployment decisions. This discrepancy is severe in clustered deployments, as existing tools either saturate network links during testing or lack proactive assessment capabilities. None of these approaches accurately predicts service co-locating compatibility in real-world scenarios. We present KUT (Konnectivity Under Test), a lightweight network assessment framework designed specifically for edge environments, combining service-specific traffic simulation with periodical monitoring to provide accurate compatibility assessments without starving co-located services. KUT enables periodic background assessments that inform orchestration decisions while consuming minimal resources.
...
Edge computing orchestration faces significant challenges due to resource constraints, highly distributed topologies, and dynamic network conditions. The discrepancy between theoretical and actual runtime performance often leads to suboptimal deployment decisions. This discrepancy is severe in clustered deployments, as existing tools either saturate network links during testing or lack proactive assessment capabilities. None of these approaches accurately predicts service co-locating compatibility in real-world scenarios. We present KUT (Konnectivity Under Test), a lightweight network assessment framework designed specifically for edge environments, combining service-specific traffic simulation with periodical monitoring to provide accurate compatibility assessments without starving co-located services. KUT enables periodic background assessments that inform orchestration decisions while consuming minimal resources.
Preprint
(2024)
-
Leonardo Tonetto, Pauline Kister, Nitinder Mohan, Jörg Ott
Networking research, especially focusing on human mobility, has evolved significantly in the last two decades and now relies on collection and analyzing larger datasets. The increasing sizes of datasets are enabled by larger automated efforts to collect data as well as by scalable methods to analyze and unveil insights, which was not possible many years ago. However, this fast expansion and innovation in human-centric research often comes at a cost of privacy or ethics. In this work, we review a vast corpus of scientific work on human mobility and how ethics and privacy were considered. We reviewed a total of 118 papers, including 149 datasets on individual mobility. We demonstrate that these ever growing collections, while enabling new and insightful studies, have not all consistently followed a pre-defined set of guidelines regarding acceptable practices in data governance as well as how their research was communicated. We conclude with a series of discussions on how data, privacy and ethics could be dealt within our community.
...
Networking research, especially focusing on human mobility, has evolved significantly in the last two decades and now relies on collection and analyzing larger datasets. The increasing sizes of datasets are enabled by larger automated efforts to collect data as well as by scalable methods to analyze and unveil insights, which was not possible many years ago. However, this fast expansion and innovation in human-centric research often comes at a cost of privacy or ethics. In this work, we review a vast corpus of scientific work on human mobility and how ethics and privacy were considered. We reviewed a total of 118 papers, including 149 datasets on individual mobility. We demonstrate that these ever growing collections, while enabling new and insightful studies, have not all consistently followed a pre-defined set of guidelines regarding acceptable practices in data governance as well as how their research was communicated. We conclude with a series of discussions on how data, privacy and ethics could be dealt within our community.
Journal article
(2024)
-
Tobias Meuser, Lauri Lovén, M Bhuyan, Shishir G. Patil, Schahram Dustdar, Atakan Aral, Suzan Bayhan, Aaron Yi Ding, Nitinder Mohan, More authors...
Edge artificial intelligence (AI) is an innovative computing paradigm that aims to shift the training and inference of machine learning models to the edge of the network. This paradigm offers the opportunity to significantly impact our everyday lives with new services such as autonomous driving and ubiquitous personalized health care. Nevertheless, bringing intelligence to the edge involves several major challenges, which include the need to constrain model architecture designs, the secure distribution and execution of the trained models, and the substantial network load required to distribute the models and data collected for training. In this article, we highlight key aspects in the development of edge AI in the past and connect them to current challenges. This article aims to identify research opportunities for edge AI, relevant to bring together the research in the fields of artificial intelligence and edge computing.
...
Edge artificial intelligence (AI) is an innovative computing paradigm that aims to shift the training and inference of machine learning models to the edge of the network. This paradigm offers the opportunity to significantly impact our everyday lives with new services such as autonomous driving and ubiquitous personalized health care. Nevertheless, bringing intelligence to the edge involves several major challenges, which include the need to constrain model architecture designs, the secure distribution and execution of the trained models, and the substantial network load required to distribute the models and data collected for training. In this article, we highlight key aspects in the development of edge AI in the past and connect them to current challenges. This article aims to identify research opportunities for edge AI, relevant to bring together the research in the fields of artificial intelligence and edge computing.
PerFail 2024
Third International Workshop on Negative Results in Pervasive Computing - Welcome and Committees
Journal article
(2024)
-
Ella Peltonen, Nitinder Mohan, Peter Zdankin, Malte Josten, Tanya Shreedar, Tanya Shreedhar, Suzan Bayhan, Javier Berrocal, Aaron Yi Ding, More authors...
Low Earth Orbit (LEO) satellite constellations have highly dynamic network topologies, making conventional routing protocols inefficient. This paper presents Geographic Checkpoint Routing (GCR), a routing protocol that combines Geographic Routing and Segment Routing (SR) principles. Utilizing the structure of Walker Delta constellations, GCR eliminates the reliance on network topologies. It routes traffic through predefined geographic segments, offloads route computation to network edges, and allows traffic engineering through customizable policies without modifying satellite infrastructure. Simulations using the Starlink constellation show that GCR can match the performance of traditional source-based routing protocols without depending on network topologies.
...
Low Earth Orbit (LEO) satellite constellations have highly dynamic network topologies, making conventional routing protocols inefficient. This paper presents Geographic Checkpoint Routing (GCR), a routing protocol that combines Geographic Routing and Segment Routing (SR) principles. Utilizing the structure of Walker Delta constellations, GCR eliminates the reliance on network topologies. It routes traffic through predefined geographic segments, offloads route computation to network edges, and allows traffic engineering through customizable policies without modifying satellite infrastructure. Simulations using the Starlink constellation show that GCR can match the performance of traditional source-based routing protocols without depending on network topologies.
It’s a bird? It’s a plane? It’s CDN!
Investigating Content Delivery Networks in the LEO Satellite Networks Era
Conference paper
(2024)
-
Rohan Bose, Saeed Fadaei, Nitinder Mohan, Mohamed Kassem, Nishanth Sastry, Jörg Ott
Content Delivery Networks (CDNs) have been pivotal in the dramatic evolution of the Internet, handling the majority of data traffic for billions of connected users. Low-Earth-Orbit (LEO) satellite networks, such as Starlink, aim to revolutionize global connectivity by providing high-speed, low-latency Internet to remote regions. However, LEO satellite networks (LSNs) face challenges integrating with traditional CDNs, which rely on geographical proximity for efficient content delivery - a method that clashes with the operational dynamics of LSNs. In this paper, we scrutinize the operation of CDNs in the context of LSNs, using Starlink as a case study. We develop a browser extension NetMet that performs extensive web browsing experiments from controlled nodes using both Starlink and terrestrial Internet access. Additionally, we analyse crowdsourced speed tests from Starlink users to Cloudflare CDN servers globally. Our results indicate significant performance issues for Starlink users, stemming from the misalignment between terrestrial and satellite infrastructures. We then investigate the potential for SpaceCDNs which integrate CDN infrastructure directly within the LSNs, and show that this approach offers a promising alternative that decreases latencies by over 50%, making them comparable with the CDN experience of users behind terrestrial ISPs. Our aim is to stimulate further research and discussion on overcoming the challenges of effective content delivery with growing LSN offerings.
...
Content Delivery Networks (CDNs) have been pivotal in the dramatic evolution of the Internet, handling the majority of data traffic for billions of connected users. Low-Earth-Orbit (LEO) satellite networks, such as Starlink, aim to revolutionize global connectivity by providing high-speed, low-latency Internet to remote regions. However, LEO satellite networks (LSNs) face challenges integrating with traditional CDNs, which rely on geographical proximity for efficient content delivery - a method that clashes with the operational dynamics of LSNs. In this paper, we scrutinize the operation of CDNs in the context of LSNs, using Starlink as a case study. We develop a browser extension NetMet that performs extensive web browsing experiments from controlled nodes using both Starlink and terrestrial Internet access. Additionally, we analyse crowdsourced speed tests from Starlink users to Cloudflare CDN servers globally. Our results indicate significant performance issues for Starlink users, stemming from the misalignment between terrestrial and satellite infrastructures. We then investigate the potential for SpaceCDNs which integrate CDN infrastructure directly within the LSNs, and show that this approach offers a promising alternative that decreases latencies by over 50%, making them comparable with the CDN experience of users behind terrestrial ISPs. Our aim is to stimulate further research and discussion on overcoming the challenges of effective content delivery with growing LSN offerings.
Poster
Twinkle, Twinkle, Streaming Star: Illuminating CDN Performance over Starlink
Low-Earth-Orbit satellite networks (LSNs) are enabling low-latency high-bandwidth internet connectivity at a global scale. However, majority of the traffic on the Internet is currently handled by Content Delivery Networks (CDNs), which rely on geographical proximity to deliver content. In this work, we examine CDN performance for the commercial largest LSN, i.e. Starlink, by performing active measurements through our web browser plugin and passive analysis of Cloudflare speed tests globally. Comparing this to terrestrial networks, we highlight significant performance degradation for Starlink users due to the asymmetries between satellite and terrestrial infrastructure.
...
Low-Earth-Orbit satellite networks (LSNs) are enabling low-latency high-bandwidth internet connectivity at a global scale. However, majority of the traffic on the Internet is currently handled by Content Delivery Networks (CDNs), which rely on geographical proximity to deliver content. In this work, we examine CDN performance for the commercial largest LSN, i.e. Starlink, by performing active measurements through our web browser plugin and passive analysis of Cloudflare speed tests globally. Comparing this to terrestrial networks, we highlight significant performance degradation for Starlink users due to the asymmetries between satellite and terrestrial infrastructure.
Conference paper
(2024)
-
Nitinder Mohan, Andrew E. Ferguson, Hendrik Cech, Rohan Bose, Prakita Rayyan Renatin, Mahesh K. Marina, Jörg Ott
The Starlink network from SpaceX stands out as the only commercial LEO network with over 2M+ customers and more than 4000 operational satellites. In this paper, we conduct a first-of-its-kind extensive multi-faceted analysis of Starlink performance leveraging several measurement sources. First, based on 19.2M crowdsourced M-Lab speed tests from 34 countries since 2021, we analyze Starlink global performance relative to terrestrial cellular networks. Second, we examine Starlink's ability to support real-time latency and bandwidth-critical applications by analyzing the performance of (i) Zoom conferencing, and (ii) Luna cloud gaming, comparing it to 5G and fiber. Third, we perform measurements from Starlink-enabled RIPE Atlas probes to shed light on the last-mile access and other factors affecting its performance.Finally, we conduct controlled experiments from Starlink dishes in two countries and analyze the impact of globally synchronized "15-second reconfiguration intervals'' of the satellite links that cause substantial latency and throughput variations. Our unique analysis paints the most comprehensive picture of Starlink's global and last-mile performance to date.
...
The Starlink network from SpaceX stands out as the only commercial LEO network with over 2M+ customers and more than 4000 operational satellites. In this paper, we conduct a first-of-its-kind extensive multi-faceted analysis of Starlink performance leveraging several measurement sources. First, based on 19.2M crowdsourced M-Lab speed tests from 34 countries since 2021, we analyze Starlink global performance relative to terrestrial cellular networks. Second, we examine Starlink's ability to support real-time latency and bandwidth-critical applications by analyzing the performance of (i) Zoom conferencing, and (ii) Luna cloud gaming, comparing it to 5G and fiber. Third, we perform measurements from Starlink-enabled RIPE Atlas probes to shed light on the last-mile access and other factors affecting its performance.Finally, we conduct controlled experiments from Starlink dishes in two countries and analyze the impact of globally synchronized "15-second reconfiguration intervals'' of the satellite links that cause substantial latency and throughput variations. Our unique analysis paints the most comprehensive picture of Starlink's global and last-mile performance to date.
Journal article
(2023)
-
Aygün Baltaci, Kaushik Chavali, Mike Kosek, Nitinder Mohan, Dominic A. Schupke, Jörg Ott
Recent industrial advancements introduce novel safety-critical applications for commercial networks. Remote Piloting (RP) Aerial Vehicles (AVs) is an example application, where reliable wireless connectivity is key to ensure safe operations in the sky. Jointly utilizing cellular and satellite networks can enable robust Multipath (MP) communications; however, their usage must be orchestrated efficiently toward application requirements. In this work, we investigate the MP communications performance of cellular and Low-Earth-Orbit (LEO) satellite links with respect to the Quality-of-Service (QoS) requirements of RP operations. Using MP-Transmission Control Protocol (MPTCP) and MP-Datagram Congestion Control Protocol (MP-DCCP), we evaluate various transport layer configurations to efficiently orchestrate both links and to support the application requirements. For this purpose, we develop an end-to-end MP emulation testbed that can provide means to realistically emulate cellular and LEO links with MPTCP and MP-DCCP. We run bi-direction al RP traffic over our testbed and measure the MP performance using different schedulers and Congestion Control (CC) algorithms. The results show that the flow size largely influences the individual path utilization due to high LEO link-layer losses. Moreover, excessive retransmissions occur on the MPTCP layer due to Head-of-Line (HoL) blocking from asymmetric link conditions. Using MP-DCCP without retransmissions helps avoid late arrivals and can meet the 99.999% communication reliability demand.
...
Recent industrial advancements introduce novel safety-critical applications for commercial networks. Remote Piloting (RP) Aerial Vehicles (AVs) is an example application, where reliable wireless connectivity is key to ensure safe operations in the sky. Jointly utilizing cellular and satellite networks can enable robust Multipath (MP) communications; however, their usage must be orchestrated efficiently toward application requirements. In this work, we investigate the MP communications performance of cellular and Low-Earth-Orbit (LEO) satellite links with respect to the Quality-of-Service (QoS) requirements of RP operations. Using MP-Transmission Control Protocol (MPTCP) and MP-Datagram Congestion Control Protocol (MP-DCCP), we evaluate various transport layer configurations to efficiently orchestrate both links and to support the application requirements. For this purpose, we develop an end-to-end MP emulation testbed that can provide means to realistically emulate cellular and LEO links with MPTCP and MP-DCCP. We run bi-direction al RP traffic over our testbed and measure the MP performance using different schedulers and Congestion Control (CC) algorithms. The results show that the flow size largely influences the individual path utilization due to high LEO link-layer losses. Moreover, excessive retransmissions occur on the MPTCP layer due to Head-of-Line (HoL) blocking from asymmetric link conditions. Using MP-DCCP without retransmissions helps avoid late arrivals and can meet the 99.999% communication reliability demand.
Journal article
(2023)
-
Ella Peltonen, Nitinder Mohan, Peter Zdankin, Tanya Shreedhar, Tri Nguyen, Suzan Bayhan, Jon Crowcroft, Jussi Kangasharju, Daniela Nicklas
Not all research leads to fruitful results; trying new ways or methods may surpass state of the art, but sometimes the hypothesis is not proven, the improvement is insignificant, or the system fails because of a design error done years ago in previous works. In a systems discipline like pervasive computing, there are many sources of errors, from hardware issues over communication channels to heterogeneous software environments. However, failure to succeed is not a failure to progress. It is essential to create platforms for sharing insights, experiences, and lessons learned when conducting research in pervasive computing so that the same mistakes are not repeated. And sometimes, a problem is a symptom of discovering new research challenges. Based on the collective input of the First International Workshop on Negative Results in Pervasive Computing (PerFail 2022), co-located with the 20th International Conference on Pervasive Computing and Communications (PerCom 2022), this article presents a comprehensive discussion on perspectives on publishing negative results, useful failures, and lessons learned in pervasive computing.
...
Not all research leads to fruitful results; trying new ways or methods may surpass state of the art, but sometimes the hypothesis is not proven, the improvement is insignificant, or the system fails because of a design error done years ago in previous works. In a systems discipline like pervasive computing, there are many sources of errors, from hardware issues over communication channels to heterogeneous software environments. However, failure to succeed is not a failure to progress. It is essential to create platforms for sharing insights, experiences, and lessons learned when conducting research in pervasive computing so that the same mistakes are not repeated. And sometimes, a problem is a symptom of discovering new research challenges. Based on the collective input of the First International Workshop on Negative Results in Pervasive Computing (PerFail 2022), co-located with the 20th International Conference on Pervasive Computing and Communications (PerCom 2022), this article presents a comprehensive discussion on perspectives on publishing negative results, useful failures, and lessons learned in pervasive computing.