arXiv subjects

Sangeetha Abdu Jyothi

Publications and source records attributed to Sangeetha Abdu Jyothi.

At least 19 recordsLinked to original sources

arXiv2026

Pramana: A Composable, Domain-Specific Backend for Empirical Networking Research

Networking research advances by turning hypotheses into empirical evidence, so accelerating it means reducing the lag between ideation (synthesizing a hypothesis) and generating the data that tests it. Consider a concrete case: does a bulk BBR download fairly share its bottleneck with competing real-time Google Meet traffic? Validating this requires configuring a realistic bottleneck link, concurrently generating BBR's bulk transfer and Meet's real-time traffic, and collecting relevant service-quality metrics. Today this overhead is high, often forcing researchers to start from scratch for every new idea. This ideation-to-data-generation gap will only worsen in the agentic AI era, where AI-assisted ideation accelerates exponentially, yet its outputs cannot be validated without a data-generation backend. This paper explores how to bridge this gap. We envision a composable, domain-specific backend, Pramana, shaped as a thin waist, with diverse research intents at the top and disparate execution substrates at the bottom. Pramana realizes this waist through a single contract, the intent specification, which disaggregates an experiment into three independent axes: the intent (what data to generate), the substrate (where to generate it), and the mechanism (how to produce it), so one specification runs on any substrate. We demonstrate Pramana's utility by building a first-of-its-kind corpus of 255 data-generation intents mined from 66 published papers, and show the intent specification satisfies all of them, where no existing tool satisfies more than 13%. Our current proof-of-concept implementation already satisfies 34% of these intents, more than twice the best existing tool, and we lay out a roadmap for closing this abstraction-implementation gap through a broader community effort to build the envisioned data-generation backend and accelerate empirical networking research.

cs.NI↗

arXiv2026

LBI: Parallel Scan Backpropagation via Latent Bounded Interfaces

Backpropagation is inherently sequential across depth, creating an $O(K)$-deep dependency chain that bottlenecks parallel training. While parallel-scan formulations theoretically reduce this depth to $O(\log K)$, they are computationally prohibitive for modern architectures due to the $O(d^3)$ cost of composing full-rank $d\times d$ Jacobians over the entire hidden state. We introduce Latent Bounded Interfaces (LBI), an algorithmic formulation that makes scan-based backpropagation tractable by restricting inter-region communication to a low-dimensional latent interface, $ m_k \in \mathbb{R}^{r}$, where $r \ll d$. This reduces the adjoint recursion to a suffix scan over $r \times r$ Jacobians, cutting per-combine cost from $O(d^3)$ to $O(r^3)$ while preserving exact gradients under the bounded-interface model. We demonstrate that LBI maintains model quality across four architectures (Mamba-2, Mamba-3, Transformer, and a Mamba--Transformer hybrid) at 47--61M block parameters. Interfaces of dimension $r=16$ suffice to preserve training quality within 0.16--0.35 cross entropy of dense baselines. The resulting framework provides an algorithmic foundation for region-parallel training, reducing cross-device backward communication to a single scan over $K$ fixed-size matrices, of approximately 56 KB for our experimental configurations.

cs.LG↗

arXiv2026

FlexiCache: Leveraging Temporal Stability of Attention Heads for Efficient KV Cache Management

Large Language Model (LLM) serving is increasingly constrained by the growing size of the key-value (KV) cache, which scales with both context length and generation length. Prior work shows that attention is dominated by a small subset of critical tokens, yet existing systems struggle to exploit this efficiently without degrading accuracy, especially in long generation. We make a key observation: the temporal stability of these critical tokens varies significantly across KV heads: some heads consistently focus on the same tokens, while others shift frequently. Building on this insight, we introduce FlexiCache, a hierarchical KV-cache management system that leverages the temporal stability of KV heads to reduce GPU memory usage and computation overhead, while preserving model accuracy. FlexiCache classifies KV heads as stable or unstable: it retains all KV-cache pages from unstable heads in GPU memory, whereas for stable heads, it keeps only the top-K pages on the GPU and offloads the rest to host memory. By exploiting temporal stability, FlexiCache performs periodic reranking for stable heads to fetch newly promoted top pages. Implemented atop vLLM, FlexiCache reduces GPU memory footprint for long-context requests by up to 70%, improves offline serving throughput by 1.38-1.55x, and lowers online token latency by 1.6-2.1x, all while maintaining accuracy in long-context, long-generation scenarios.

cs.LG↗

arXiv2026

Airavat: An Agentic Framework for Internet Measurement

Internet measurement faces twin challenges: complex analyses require expert-level orchestration of tools, yet even syntactically correct implementations can have methodological flaws and can be difficult to verify. Democratizing measurement capabilities thus demands automating both workflow generation and verification against methodological standards established through decades of research. We present Airavat, the first agentic framework for Internet measurement workflow generation with systematic verification and validation. Airavat coordinates a set of agents mirroring expert reasoning: three agents handle problem decomposition, solution design, and code implementation, with assistance from a registry of existing tools. Two specialized engines ensure methodological correctness: a Verification Engine evaluates workflows against a knowledge graph encoding five decades of measurement research, while a Validation Engine identifies appropriate validation techniques grounded in established methodologies. Through four Internet measurement case studies, we demonstrate that Airavat (i) generates workflows matching expert-level solutions, (ii) makes sound architectural decisions, (iii) addresses novel problems without ground truth, and (iv) identifies methodological flaws missed by standard execution-based testing.

cs.NI↗

arXiv2025

CoopQ: Cooperative Game Inspired Layerwise Mixed Precision Quantization for LLMs

Large Language Models (LLMs) promise impressive capabilities, yet their multi-billion-parameter scale makes on-device or low-resource deployment prohibitive. Mixed-precision quantization offers a compelling solution, but existing methods struggle when the average precision drops below four bits, as they rely on isolated, layer-specific metrics that overlook critical inter-layer interactions affecting overall performance. To address these limitations, we first frame the mixed-precision quantization problem as a cooperative game among layers and introduce Shapley-based Progressive Quantization Estimation (SPQE) to efficiently obtain accurate Shapley estimates of layer sensitivities and inter-layer interactions. Leveraging the SPQE estimates, we propose Cooperative Game Inspired Mixed-Precision Quantization (CoopQ) which translates these Shapley estimates into a binary quadratic optimization formulation, assigning either 2 or 4-bit precision to layers under strict memory constraints. Comprehensive experiments conducted on Llama-3, Gemma-2, and Qwen-3 models across three independent PTQ backends (Quanto, HQQ, GPTQ) demonstrate CoopQ's scalability and consistently superior performance compared to methods relying solely on isolated metrics. Across average precisions spanning 4 bit down to 2 bit, CoopQ cuts Perplexity by 20 - 80 % relative to the best baseline, with the margin growing as the bit-width tightens.

cs.LG↗

arXiv2025

Towards an Agentic Workflow for Internet Measurement Research

Internet measurement research faces an accessibility crisis: complex analyses require custom integration of multiple specialized tools that demands specialized domain expertise. When network disruptions occur, operators need rapid diagnostic workflows spanning infrastructure mapping, routing analysis, and dependency modeling. However, developing these workflows requires specialized knowledge and significant manual effort. We present ArachNet, the first system demonstrating that LLM agents can independently generate measurement workflows that mimics expert reasoning. Our core insight is that measurement expertise follows predictable compositional patterns that can be systematically automated. ArachNet operates through four specialized agents that mirror expert workflow, from problem decomposition to solution implementation. We validate ArachNet with progressively challenging Internet resilience scenarios. The system independently generates workflows that match expert-level reasoning and produce analytical outputs similar to specialist solutions. Generated workflows handle complex multi-framework integration that traditionally requires days of manual coordination. ArachNet lowers barriers to measurement workflow composition by automating the systematic reasoning process that experts use, enabling broader access to sophisticated measurement capabilities while maintaining the technical rigor required for research-quality analysis.

cs.NI↗

arXiv2025

AMAQ: Adaptive Mixed-bit Activation Quantization for Collaborative Parameter Efficient Fine-tuning

Large Language Models (LLMs) are scaling rapidly, creating significant challenges for collaborative server client distributed training, particularly in terms of communication efficiency and computational overheads. To address these challenges, we implement Parameter-efficient Split Learning, which effectively balances efficiency and performance for collaborative training on low-resource devices. To reduce communication overhead in collaborative training, we introduce Adaptive Mixed bit Activation Quantization (AMAQ), a strategy that progressively compresses activations and gradients from high precision (6 to 8 bits) to low precision (3 to 4 bits). AMAQ achieves this by effectively allocating bit budgets across channels based on feature wise and layer wise importance using bit regularization. Under the same bit budgets, AMAQ outperforms fixed-precision approaches, delivering about 2.5% higher generation accuracy and about 1.3% better classification accuracy for models like LLaMA3 8B and Qwen2.5 7B. In addition, it significantly enhances training stability and reducing ultra-low bit representation collapse during the training. Experiments demonstrate that AMAQ integrates effectively into practical multi-machine collaborative training setups, offering superior inference accuracy with only a modest communication overhead for bits adaptation during training. This trade off makes AMAQ a practical and effective solution for collaborative training with minimal communication cost.

cs.LG↗

arXiv2025

A Deep Dive into the Impact of Solar Storms on LEO Satellite Networks

Low Earth Orbit (LEO) satellite networks are an important part of the global communication infrastructure today. Despite ongoing efforts to improve their resilience, they remain vulnerable to component damage and deorbiting under harsh space weather conditions. Prior work identified a modest but noticeable impact on LEO satellite network performance during solar storms, typically manifesting as an immediate rise in packet loss and a sustained increase in round-trip time (RTT). However, these studies offer only coarse-grained insights and do not capture the nuanced spatial and temporal patterns of disruption across the LEO network. In this paper, we conduct a deep dive into the impact of solar storms on LEO satellite communications. By localizing the impact of increased atmospheric drag at the level of individual satellites and orbits, we reveal significant heterogeneity in how different parts of the network are affected. We find that the degree of performance degradation varies significantly across geographic regions, depending on satellite positioning during the storm. Specifically, we find that (i) not all satellite orbits are equally vulnerable, (ii) within a given orbit, certain satellites experience disproportionate impact depending on their position relative to geomagnetic conditions, and (iii) autonomous maneuvering of satellites might be a cause of the sustained increase in RTT. Our findings uncover previously overlooked patterns of vulnerability in LEO satellite constellations and highlight the need for more adaptive, region-aware mitigation strategies to address space weather-induced network disruptions.

astro-ph.EP↗

arXiv2025

ShareLoRA: Parameter Efficient and Robust Large Language Model Fine-tuning via Shared Low-Rank Adaptation

In this paper, we introduce \textbf{Share}d \textbf{Lo}w \textbf{R}ank \textbf{A}daptation (ShareLoRA), a Large Language Model (LLM) fine-tuning technique that balances parameter efficiency, adaptability, and robustness without compromising performance. By strategically sharing the low-rank weight matrices across different layers, ShareLoRA achieves 44\% to 96\% reduction in trainable parameters compared to standard LoRA, alongside a substantial decrease in memory overhead. This efficiency gain scales with model size, making ShareLoRA particularly advantageous for resource-constrained environments. Importantly, ShareLoRA not only maintains model performance but also exhibits robustness in both classification and generation tasks across diverse models, including RoBERTa, GPT-2, and LLaMA series (1, 2, and 3). It consistently outperforms LoRA in zero-shot, few-shot, and continual fine-tuning scenarios, achieving up to 1.2\% average accuracy improvement, and enhanced generalization across domains. In continual learning settings, ShareLoRA achieves 1.2\% higher accuracy on GSM8K, 0.6\% on HumanEval, and 0.5\% on both MMLU and MMLU-Pro. Our results demonstrate that ShareLoRA supports high-quality fine-tuning while offering strong generalization and continual adaptation across various model scales and diverse tasks.

cs.CL↗

arXiv2025

Leveraging Traceroute Inconsistencies to Improve IP Geolocation

Traceroutes and geolocation are two essential network measurement tools that aid applications such as network mapping, topology generation, censorship, and Internet path analysis. However, these tools, individually and when combined, have significant limitations that can lead to inaccurate results. Prior research addressed specific issues with traceroutes and geolocation individually, often requiring additional measurements. In this paper, we introduce GeoTrace, a lightweight tool designed to identify, classify, and resolve geolocation anomalies in traceroutes using existing data. GeoTrace leverages the abundant information in traceroutes and geolocation databases to identify anomalous IP addresses with incorrect geolocation. It systematically classifies these anomalies based on underlying causes - such as MPLS effects or interface discrepancies - and refines their geolocation estimates where possible. By correcting these inaccuracies, GeoTrace enhances the reliability of traceroute-based analyses without the need for additional probing. Our work offers a streamlined solution that enhances the accuracy of geolocation in traceroute analysis, paving the way for more reliable measurement studies.

cs.NI↗

arXiv2024

Cooperative Graceful Degradation In Containerized Clouds

Cloud resilience is crucial for cloud operators and the myriad of applications that rely on the cloud. Today, we lack a mechanism that enables cloud operators to perform graceful degradation of applications while satisfying the application's availability requirements. In this paper, we put forward a vision for automated cloud resilience management with cooperative graceful degradation between applications and cloud operators. First, we investigate techniques for graceful degradation and identify an opportunity for cooperative graceful degradation in public clouds. Second, leveraging criticality tags on containers, we propose diagonal scaling -- turning off non-critical containers during capacity crunch scenarios -- to maximize the availability of critical services. Third, we design Phoenix, an automated cloud resilience management system that maximizes critical service availability of applications while also considering operator objectives, thereby improving the overall resilience of the infrastructure during failures. We experimentally show that the Phoenix controller running atop Kubernetes can improve critical service availability by up to $2\times$ during large-scale failures. Phoenix can handle failures in a cluster of 100,000 nodes within 10 seconds. We also develop AdaptLab, an open-source resilience benchmarking framework that can emulate realistic cloud environments with real-world application dependency graphs.

cs.NI↗

arXiv2024

An In-Depth Investigation of LEO Satellite Topology Design Parameters

Low Earth Orbit (LEO) satellite networks are rapidly gaining traction today. Although several real-world deployments exist, our preliminary analysis of LEO topology performance with the soon-to-be operational Inter-Satellite Links (ISLs) reveals several interesting characteristics that are difficult to explain based on our current understanding of topologies. For example, a real-world satellite shell with a low density of satellites offers better latency performance than another shell with nearly double the number of satellites. In this work, we conduct an in-depth investigation of LEO satellite topology design parameters and their impact on network performance while using the ISLs. In particular, we focus on three design parameters: the number of orbits in a shell, the inclination of orbits, and the number of satellites per orbit. Through an extensive analysis of real-world and synthetic satellite configurations, we uncover several interesting properties of satellite topologies. Notably, there exist thresholds for the number of satellites per orbit and the number of orbits below which the latency performance degrades significantly. Moreover, network delay between a pair of traffic endpoints depends on the alignment of the satellite's orbit (Inclination) with the geographic locations of endpoints.

cs.NI↗

arXiv2024

CrystalBox: Future-Based Explanations for Input-Driven Deep RL Systems

We present CrystalBox, a novel, model-agnostic, posthoc explainability framework for Deep Reinforcement Learning (DRL) controllers in the large family of input-driven environments which includes computer systems. We combine the natural decomposability of reward functions in input-driven environments with the explanatory power of decomposed returns. We propose an efficient algorithm to generate future-based explanations across both discrete and continuous control environments. Using applications such as adaptive bitrate streaming and congestion control, we demonstrate CrystalBox's capability to generate high-fidelity explanations. We further illustrate its higher utility across three practical use cases: contrastive explanations, network observability, and guided reward design, as opposed to prior explainability techniques that identify salient features.

cs.LG↗

arXiv2024

Xaminer: An Internet Cross-Layer Resilience Analysis Tool

A resilient Internet infrastructure is critical in our highly interconnected society. However, the Internet faces several vulnerabilities, ranging from natural disasters to human activities, that can impact the physical layer and, in turn, the higher network layers, such as IP links. In this paper, we introduce Xaminer, the first Internet cross-layer resilience analysis tool, to evaluate the interplay between physical- and network-layer failures. Using a cross-layer Internet map and a failure event model, Xaminer generates a risk profile encompassing a cross-layer impact report, critical infrastructure identification at each layer, and the discovery of trends and patterns under different failure event settings. Xaminer's key strengths lie in its adaptability to diverse disaster scenarios, the ability to assess risks at various granularities, and the capability to generate joint risk profiles for multiple events. We demonstrate Xaminer's capabilities in cross-layer analysis across a spectrum of disaster event models and regions, showcasing its potential role in facilitating well-informed decision-making for resilience planning and deployments.

cs.NI↗

arXiv2023

LinguaLinked: A Distributed Large Language Model Inference System for Mobile Devices

Deploying Large Language Models (LLMs) locally on mobile devices presents a significant challenge due to their extensive memory requirements. In this paper, we introduce LinguaLinked, a system for decentralized, distributed LLM inference on mobile devices. LinguaLinked enables collaborative execution of the inference task across multiple trusted devices. LinguaLinked ensures data privacy by processing information locally. LinguaLinked uses three key strategies. First, an optimized model assignment technique segments LLMs and uses linear optimization to align segments with each device's capabilities. Second, an optimized data transmission mechanism ensures efficient and structured data flow between model segments while also maintaining the integrity of the original model structure. Finally, LinguaLinked incorporates a runtime load balancer that actively monitors and redistributes tasks among mobile devices to prevent bottlenecks, enhancing the system's overall efficiency and responsiveness. We demonstrate that LinguaLinked facilitates efficient LLM inference while maintaining consistent throughput and minimal latency through extensive testing across various mobile devices, from high-end to low-end Android devices. In our evaluations, compared to the baseline, LinguaLinked achieves an inference performance acceleration of $1.11\times$ to $1.61\times$ in single-threaded settings, $1.73\times$ to $2.65\times$ with multi-threading. Additionally, runtime load balancing yields an overall inference acceleration of $1.29\times$ to $1.32\times$.

cs.LG↗

arXiv2023

Plume: A Framework for High Performance Deep RL Network Controllers via Prioritized Trace Sampling

Deep Reinforcement Learning (DRL) has shown promise in various networking environments. However, these environments present several fundamental challenges for standard DRL techniques. They are difficult to explore and exhibit high levels of noise and uncertainty. Although these challenges complicate the training process, we find that in practice we can substantially mitigate their effects and even achieve state-of-the-art real-world performance by addressing a factor that has been previously overlooked: the skewed input trace distribution in DRL training datasets. We introduce a generalized framework, Plume, to automatically identify and balance the skew using a three-stage process. First, we identify the critical features that determine the behavior of the traces. Second, we classify the traces into clusters. Finally, we prioritize the salient clusters to improve the overall performance of the controller. Plume seamlessly works across DRL algorithms, without requiring any changes to the DRL workflow. We evaluated Plume on three networking environments, including Adaptive Bitrate Streaming, Congestion Control, and Load Balancing. Plume offers superior performance in both simulation and real-world settings, across different controllers and DRL algorithms. For example, our novel ABR controller, Gelato trained with Plume consistently outperforms prior state-of-the-art controllers on the live streaming platform Puffer for over a year. It is the first controller on the platform to deliver statistically significant improvements in both video quality and stalling, decreasing stalls by as much as 75%.

cs.LG↗

arXiv2023

Nautilus: A Framework for Cross-Layer Cartography of Submarine Cables and IP Links

Submarine cables constitute the backbone of the Internet. However, these critical infrastructure components are vulnerable to several natural and man-made threats, and during failures, are difficult to repair in their remote oceanic environments. In spite of their crucial role, we have a limited understanding of the impact of submarine cable failures on global connectivity, particularly on the higher layers of the Internet. In this paper, we present Nautilus, a framework for cross-layer cartography of submarine cables and IP links. Using a corpus of public datasets and Internet cartographic techniques, Nautilus identifies IP links that are likely traversing submarine cables and maps them to one or more potential cables. Nautilus also gives each IP to cable assignment a prediction score that reflects the confidence in the mapping. Nautilus generates a mapping for 3.05 million and 1.43 million IPv4 and IPv6 links respectively, covering 91% of all active cables. In the absence of ground truth data, we validate Nautilus mapping using three techniques: analyzing past cable failures, using targeted traceroute measurements, and comparing with public network maps of two operators.

cs.NI↗

arXiv2023

Characterizing the Role of Power Grids in Internet Resilience

Among critical infrastructures, power grids and communication infrastructure are identified as uniquely critical since they enable the operation of all other sectors. Due to their vital role, the research community has undertaken extensive efforts to understand the complex dynamics and resilience characteristics of these infrastructures, albeit independently. However, power and communication infrastructures are also interconnected, and the nature of the Internet's dependence on power grids is poorly understood. In this paper, we take the first step toward characterizing the role of power grids in Internet resilience by analyzing the overlap of global power and Internet infrastructures. We investigate the impact of power grid failures on Internet availability and find that nearly $65\%$ of the public Internet infrastructure components are concentrated in a few ($< 10$) power grid failure zones. More importantly, power grid dependencies severely limit the number of disjoint availability zones of cloud providers. When dependency on grids serving data center locations is taken into account, the number of isolated AWS Availability Zones reduces from 87 to 19. Building upon our findings, we develop NetWattZap, an Internet resilience analysis tool that generates power grid dependency-aware deployment suggestions for Internet infrastructure and application components, which can also take into account a wide variety of user requirements.

cs.NI↗