About

Venkat Arun is an Assistant Professor at the University of Texas at Austin in the Department of Computer Science. His research focuses on Formal Methods, Systems, and Networking. He is involved in advancing the understanding and development of formal verification techniques, system reliability, and network security. His work contributes to creating more robust and secure computing systems by applying rigorous mathematical and logical methods to software and hardware design. Venkat Arun's background includes significant contributions to the field of formal methods, emphasizing the development of tools and techniques for verifying the correctness of complex systems. His research aims to improve the dependability of software and hardware systems, ensuring they operate correctly under various conditions. Through his academic and research activities, he collaborates with other experts to push the boundaries of system reliability and security, impacting both theoretical foundations and practical applications in computer science.

Research topics

• Computer Science
• Artificial Intelligence
• Mathematics
• Statistics
• Mathematical optimization
• Geography
• Physics
• Telecommunications
• Distributed computing
• Acoustics
• Computer network
• Mechanical engineering
• Theoretical computer science
• Engineering

Selected publications

• Bring Your Own Objective: Inter-operability of Network Objectives in Datacenters

  ArXiv.org · 2026-02-10

  articleOpen access

  Datacenter networks are currently locked in a "tyranny of the single objective". While modern workloads demand diverse performance goals, ranging from coflow completion times, per-flow fairness, short-flow latencies, existing fabrics are typically hardcoded for a single metric. This rigid coupling ensures peak performance when application and network objectives align, but results in abysmal performance when they diverge. We propose DMart, a decentralized scheduling framework that treats network bandwidth as a competitive marketplace. In DMart, applications independently encode the urgency and importance of their network traffic into autonomous bids, allowing diverse objectives to co-exist natively on the same fabric. To meet the extreme scale and sub-microsecond requirements of modern datacenters, DMart implements distributed, per-link, per-RTT auctions, without relying on ILPs, centralized schedulers, or complex priority queues. We evaluate DMart using packet-level simulations and compare it against network schedulers designed for individual metrics, e.g., pFabric and Sincronia. DMart matches the performance of specialized schedulers on their own "home turf" while simultaneously optimizing secondary metrics. Compared to pFabric and Sincronia, DMart reduces deadline misses by 2x and coflow completion times by 1.6x respectively, while matching pFabric short-flow completion times.

  PublisherOA PDF

• Bring Your Own Objective: Inter-operability of Network Objectives in Datacenters

  Open MIND · 2026-02-10

  preprint

  Datacenter networks are currently locked in a "tyranny of the single objective". While modern workloads demand diverse performance goals, ranging from coflow completion times, per-flow fairness, short-flow latencies, existing fabrics are typically hardcoded for a single metric. This rigid coupling ensures peak performance when application and network objectives align, but results in abysmal performance when they diverge. We propose DMart, a decentralized scheduling framework that treats network bandwidth as a competitive marketplace. In DMart, applications independently encode the urgency and importance of their network traffic into autonomous bids, allowing diverse objectives to co-exist natively on the same fabric. To meet the extreme scale and sub-microsecond requirements of modern datacenters, DMart implements distributed, per-link, per-RTT auctions, without relying on ILPs, centralized schedulers, or complex priority queues. We evaluate DMart using packet-level simulations and compare it against network schedulers designed for individual metrics, e.g., pFabric and Sincronia. DMart matches the performance of specialized schedulers on their own "home turf" while simultaneously optimizing secondary metrics. Compared to pFabric and Sincronia, DMart reduces deadline misses by 2x and coflow completion times by 1.6x respectively, while matching pFabric short-flow completion times.

  DOI

• Faster-than-light coordination for networked systems with quantum non-local games

  2025-11-17

  articleOpen access1st authorCorresponding

  Many networked systems rely on hashing and randomized algorithms for tasks such as load balancing, thereby avoiding the need for coordination or communication among participants on each request. However, purely random routing can lead to collisions and missed opportunities for beneficial colocation. Quantum entanglement enables participants to instantly make correlated decisions without communicating. We explore how this capability can expand the Pareto frontier of achievable performance in networked systems, presenting both positive and negative results. Notably, many of these advantages can be realized using small, currently available quantum devices that can often operate at room temperature.

  PublisherDOI

• Compact 9-Level Double Boost Inverter (C9LDBI) with SPWM Control

  2025-07-23

  articleSenior author

  This article presents a modular design and control strategy for the Compact 9 level Double Boost Inverter (C9LDBI) when used with Sinusoidal pulse width modulation (SPWM). This approach aims to address the drawbacks of the conventional Multilevel inverters, such as the need for more switching, diodes and capacitors. The suggested C9LDBI can do double voltage boost nine level output using one DC supply, three capacitors, nine switches and three diodes. Conventional inverters take more parts than C9LDBI. In addition, the multi-carrier modulation system with SPWM control is helping to produce better output waveform. MATLAB simulation test findings about the attributes of lower Total Harmonic Distortion and effectively high DC bus utilisation suggest the general effectiveness of C9LDBI under various of SPWM control with the modulation index variation.

  PublisherDOI

• IoT based Harbour Surveillance Aqua Guard RC Canoe

  2025-04-07

  article

  India’s harbour ecosystems face significant environmental and security risks that require constant attention. The current surveillance system involves manual supervision and less than basic threshold alerted detection to critical events with limited integration of harbour environmental monitoring. Thus, this thesis proposes an IoT anomaly detection system that amalgamates environmental monitoring and real-time surveillance monitoring management. Aqua Guard RC Canoe is proposed to monitor the harbour area that can generate data for anomalies and threats present due to environmental sensors like pH, turbidity, anemometer, temperature, humidity, ESP32 camera, and GPS. It delivers the seamless interconnectivity of sensory devices, transmission, and integration on the IoT network. When an anomaly is detected, the system will be configured to raise alarms and generate visual evidence to initiate a swift response mechanism to the anomaly. For flexible and real-time monitoring of various spots in the harbour, it will be mounted on a remote-controlled mobile boat. A more proactive flow of decisions for risk mitigation is expected from the aggregation and surveillance data as well as environmental parameters to get a balanced view of the conditions of the harbour.

  PublisherDOI

• Robust Heuristic Algorithm Design with LLMs

  ArXiv.org · 2025-10-09

  preprintOpen access

  We posit that we can generate more robust and performant heuristics if we augment approaches using LLMs for heuristic design with tools that explain why heuristics underperform and suggestions about how to fix them. We find even simple ideas that (1) expose the LLM to instances where the heuristic underperforms; (2) explain why they occur; and (3) specialize design to regions in the input space, can produce more robust algorithms compared to existing techniques~ -- ~the heuristics we produce have a $\sim28\times$ better worst-case performance compared to FunSearch, improve average performance, and maintain the runtime.

  PublisherOA PDFDOI

• Lightweight Automated Reasoning for Network Architectures

  2024-11-11 · 2 citations

  articleOpen access

  Architecting a modern data center network is increasingly complicated. Seeking the highest performance and support for emerging workloads, network architects planning a buildout must choose from a large selection of switching components, NICs, network stacks, congestion control algorithms, routing schemes, measurement systems, virtualization software, centralized bandwidth allocators and security mechanisms, all from various vendors. Today, manual planning by human experts is time-consuming at best, and can easily result in overlooked design choices or missed complex inter-dependencies.

  PublisherDOI

• Enhancing energy hub management with unified plug-in electric vehicle based demand response and energy storage systems

  Journal of Energy Storage · 2024-12-19 · 10 citations

  articleSenior author
  PublisherDOI

• A Performance Verification Methodology for Resource Allocation Heuristics

  arXiv (Cornell University) · 2023-01-10

  preprintOpen access

  Performance verification is a nascent but promising tool for understanding the performance and limitations of heuristics under realistic assumptions. Bespoke performance verification tools have already demonstrated their value in settings like congestion control and packet scheduling. In this paper, we aim to emphasize the broad applicability and utility of performance verification. To that end, we highlight the design principles of performance verification. Then, we leverage that understanding to develop a set of easy-to-follow guidelines that are applicable to a wide range of resource allocation heuristics. In particular, we introduce Virelay, a framework that enables heuristic designers to express the behavior of their algorithms and their assumptions about the system in an environment that resembles a discrete-event simulator. We demonstrate the utility and ease-of-use of Virelay by applying it to six diverse case studies. We produce bounds on the performance of classical algorithms, work stealing and SRPT scheduling, under practical assumptions. We demonstrate Virelay's expressiveness by capturing existing models for congestion control and packet scheduling, and we verify the observation that TCP unfairness can cause some ML training workloads to spontaneously converge to a state of high network utilization. Finally, we use Virelay to identify two bugs in the Linux CFS load balancer.

  PublisherOA PDFDOI

• Image-based Driver Alert System for Prevention of Fatigue-related Accidents

  E3S Web of Conferences · 2023-01-01

  articleOpen accessSenior author

  The objective of this project is to design a driver unconsciousness detection system using image processing to detect drowsiness and unconsciousness in drivers, thereby preventing accidents resulting from driver fatigue. Driver fatigue is a serious road safety issue, with approximately 20% of all road accidents attributed to this cause. Conventional drowsiness detection systems rely on physiological monitoring, which can be unreliable, expensive, and challenging to implement and maintain. In contrast, the proposed system monitors a sequence of images to identify facial and behavioral patterns indicative of drowsiness or unconsciousness. By detecting facial landmark points and analyzing the duration of eye closure, the system can accurately classify the driver’s state and take appropriate measures such as reducing the vehicle’s speed and alerting emergency services of the driver’s geo-location. The successful implementation of this system holds immense potential for substantially reducing the number of accidents resulting from driver fatigue, thereby mitigating the loss of lives and injuries.

  PublisherOA PDFDOI

Frequent coauthors

• Ahmed Saeed

  Georgia Institute of Technology

  5 shared

• Hari Balakrishnan

  IIT@MIT

  4 shared

• Radhika Mittal

  3 shared

• Hamsa Balakrishnan

  American Institute of Aeronautics and Astronautics

  3 shared

• S. V. S. Prasad

  3 shared

• Mohammad Reza Alizadeh

  3 shared

• Arjun Balasingam

  3 shared

• K. Nishanth Rao

  2 shared

Awards & honors

• best paper awards
• Marconi Society young scholar award