arXiv subjects

Praveen Tammana

Publications and source records attributed to Praveen Tammana.

6 recordsLinked to original sources

arXiv2026

Simulating Unified Tensor Resharding in heterogeneous AI systems

State-of-the-art AI training simulators assume homogeneous compute and network infrastructure. However, real-world training infrastructure is becoming increasingly heterogeneous since: (a) Model architectures such as multimodal and MoE exploit heterogeneity to improve device utilization, (b) Public cloud platforms often provide limited availability of homogeneous hardware due to fast hardware evolution, and (c) Large enterprises frequently deploy geographically distributed infrastructure that is both diverse and heterogeneous. In this paper, we present Xsim, a heterogeneity-aware simulator for distributed LLM training. Xsim supports: (i) Load balancing through non-uniform workload partitioning across heterogeneous device groups, (ii) Heterogeneity-aware collective communication via customized ring construction and chunk partitioning, (iii) Reusable heterogeneity-aware abstractions for emerging pipeline-parallel algorithms and non-uniform tensor resharding technique, (iv) Flexible input abstractions for specifying deployment plans with custom device groups and custom device-to-parallelism mappings, and (v) Pluggable integration with NS-3 and htsim, allowing users to trade off simulation fidelity for performance and scalability. Our evaluation demonstrates that Xsim accurately predicts training time for real-world heterogeneous deployments, with an error of less than 5% across most heterogeneous data-parallel/tensor-parallel configurations and around 2% error with pipeline-parallel communication modeling. We expose actionable metrics such as pipeline bubble time and straggler waiting time.

cs.DC↗

arXiv2026

Yaksha-Prashna: Understanding eBPF Bytecode Network Function Behavior

Many cloud infrastructure organizations increasingly rely on third-party eBPF-based network functions for use cases like security, observability, and load balancing, so that not everyone requires a team of highly skilled eBPF experts. However, the network functions from third parties (e.g., F5, Palo Alto) are available in bytecode format to cloud operators, giving little or no understanding of their functional correctness and interaction with other network functions in a chain. Also, eBPF developers want to provide proof of functional correctness for their developed network functions without disclosing the source code to the operators. We design Yaksha-Prashna, a system that allows operators/developers to assert and query bytecode's conformance to its specification and dependencies on other bytecodes. Our work builds domain-specific models that enable us to employ scalable program analysis to extract and model eBPF programs. Using Yaksha-Prashna language, we express 24 properties on standard and non-standard eBPF-based network functions with 200-1000x speedup over the state-of-the-art work.

cs.CR↗

arXiv2025

Simulating LLM training workloads for heterogeneous compute and network infrastructure

The growing demand for large-scale GPU clusters in distributed model training presents a significant barrier to innovation, particularly in model optimization, performance tuning, and system-level enhancements. To address this challenge, LLM training simulators are employed to estimate training time and guide design decisions. However, the state-of-the-art LLM training simulators assume homogeneous compute and network infrastructure. In practice, device heterogeneity is inevitable due to resource sharing in cloud environments, frequent shifts in device generations, and inherent intra-chip interconnect heterogeneity. To address the gap between state-of-the-art and practical requirements, we propose the design of a heterogeneity-aware distributed LLM simulator capable of predicting training time while enabling abstractions to specify custom configurations for device groups and device-to-parallelism mapping. We present the design requirements and challenges in building a heterogeneity-aware distributed ML training simulator, and design components such as non-uniform workload partitioning. Our initial simulation results demonstrate the impact of heterogeneity on the model computation and communication time.

cs.DC↗

arXiv2025

GPUs, CPUs, and... NICs: Rethinking the Network's Role in Serving Complex AI Pipelines

The increasing prominence of AI necessitates the deployment of inference platforms for efficient and effective management of AI pipelines and compute resources. As these pipelines grow in complexity, the demand for distributed serving rises and introduces much-dreaded network delays. In this paper, we investigate how the network can instead be a boon to the excessively high resource overheads of AI pipelines. To alleviate these overheads, we discuss how resource-intensive data processing tasks -- a key facet of growing AI pipeline complexity -- are well-matched for the computational characteristics of packet processing pipelines and how they can be offloaded onto SmartNICs. We explore the challenges and opportunities of offloading, and propose a research agenda for integrating network hardware into AI pipelines, unlocking new opportunities for optimization.

cs.NI↗

arXiv2019

Contra: A Programmable System for Performance-aware Routing

We present Contra, a system for performance-aware routing that can adapt to traffic changes at hardware speeds. While existing work has developed point solutions for performance-aware routing on a fixed topology (e.g., a Fattree) with a fixed routing policy (e.g., use least utilized paths), Contra can be configured to operate seamlessly over any network topology and a wide variety of sophisticated routing policies. Users of Contra write network-wide policies that rank network paths given their current performance. A compiler then analyzes such policies in conjunction with the network topology and decomposes them into switch-local P4 programs, which collectively implement a new, specialized distance-vector protocol. This protocol generates compact probes that traverse the network, gathering path metrics to optimize for the user policy dynamically. Switches respond to changing network conditions at hardware speeds by routing flowlets along the best policy-compliant paths. Our experiments show that Contra scales to large networks, and that in terms of flow completion times, it is competitive with hand-crafted systems that have been customized for specific topologies and policies.

cs.NI↗

arXiv2017

Fault Localization in Large-Scale Network Policy Deployment

The recent advances in network management automation and Software-Defined Networking (SDN) are easing network policy management tasks. At the same time, these new technologies create a new mode of failure in the management cycle itself. Network policies are presented in an abstract model at a centralized controller and deployed as low-level rules across network devices. Thus, any software and hardware element in that cycle can be a potential cause of underlying network problems. In this paper, we present and solve a network policy fault localization problem that arises in operating policy management frameworks for a production network. We formulate our problem via risk modeling and propose a greedy algorithm that quickly localizes faulty policy objects in the network policy. We then design and develop SCOUT---a fully-automated system that produces faulty policy objects and further pinpoints physical-level failures which made the objects faulty. Evaluation results using a real testbed and extensive simulations demonstrate that SCOUT detects faulty objects with small false positives and false negatives.

cs.NI↗