arXiv subjects

Parv Agarwal

Publications and source records attributed to Parv Agarwal.

4 recordsLinked to original sources

arXiv2026

GPUAlert: A Zero-Instrumentation Process-Boundary Monitor for Diagnosing GPU Training-Job Failures

GPU training jobs fail often, roughly two in five on large production clusters, yet the operator typically learns of a failure only by reconnecting hours later. Experiment trackers require editing the training script and maintaining a cloud connection; the scheduler's mail hook delivers a single status line with no cause and no logs. GPUAlert is a command-line wrapper that monitors any training command at the process boundary, and with no change to that command, emails a structured notification on completion carrying a classified failure cause, durable logs, and output artifacts. The tool is organized around three reliability primitives: a pre-launch log guarantee that establishes the durable destination before the child process can crash, notifier isolation that makes the wrapper's exit code a pure function of the child's status regardless of whether the email succeeds, and a non-silent artifact budget that bounds attachment size without ever dropping output silently. We release a labelled corpus of 474 GPU training logs across 15 failure classes and a reproducible evaluation harness. On the twelve hardware-reproduced classes, the ordered-rule classifier reaches 0.997 macro-F1, against 0.830 for unordered keyword matching and 0.133 for exit-code inspection. Wrapper overhead is a constant approximately 3ms per job; the pre-launch guarantee preserves a log where a shell redirect yields nothing; and across all 15 failure modes the wrapper returns the child's exit code unchanged even when the SMTP relay is unreachable.

cs.SE↗

arXiv2026

RIPPLE++: An Incremental Framework for Efficient GNN Inference on Evolving Graphs

Real-world graphs are dynamic, with frequent updates to their structure and features due to evolving vertex and edge properties. These continual changes pose significant challenges for efficient inference in graph neural networks (GNNs). Existing vertex-wise and layer-wise inference approaches are ill-suited for dynamic graphs, as they incur redundant computations, large neighborhood traversals, and high communication costs, especially in distributed settings. Additionally, while sampling-based approaches can be adopted to approximate final layer embeddings, these are often not preferred in critical applications due to their non-determinism. These limitations hinder low-latency inference required in real-time applications. To address this, we propose RIPPLE++, a framework for streaming GNN inference that efficiently and accurately updates embeddings in response to changes in the graph structure or features. RIPPLE++ introduces a generalized incremental programming model that captures the semantics of GNN aggregation functions and incrementally propagates updates to affected neighborhoods. RIPPLE++ accommodates all common graph updates, including vertex/edge addition/deletions and vertex feature updates. RIPPLE++ supports both single-machine and distributed deployments. On a single machine, it achieves up to $56$K updates/sec on sparse graphs like Arxiv ($169$K vertices, $1.2$M edges), and about $7.6$K updates/sec on denser graphs like Products ($2.5$M vertices, $123.7$M edges), with latencies of $0.06$--$960$ms, and outperforming state-of-the-art baselines by $2.2$--$24\times$ on throughput. In distributed settings, RIPPLE++ offers up to $\approx25\times$ higher throughput and $20\times$ lower communication costs compared to recomputing baselines.

cs.DC↗

arXiv2024

Motif Caller: Sequence Reconstruction for Motif-Based DNA Storage

DNA data storage is rapidly emerging as a promising solution for long-term data archiving, largely due to its exceptional durability. However, the synthesis of DNA strands remains a significant bottleneck in terms of cost and speed. To address this, new methods have been developed that encode information by concatenating long data-carrying DNA sequences from pre-synthesized DNA subsequences - known as motifs - from a library. Reading back data from DNA storage relies on basecalling - the process of translating raw nanopore sequencing signals into DNA base sequences using machine learning models. These sequences are then decoded back into binary data. However, current basecalling approaches are not optimized for decoding motif-carrying DNA: they first predict individual bases from the raw signal and only afterward attempt to identify higher-level motifs. This two-step, motif-agnostic process is both imprecise and inefficient. In this paper we introduce Motif Caller, a machine learning model designed to directly detect entire motifs from raw nanopore signals, bypassing the need for intermediate basecalling. By targeting motifs directly, Motif Caller leverages richer signal features associated with each motif, resulting in significantly improved accuracy. This direct approach also enhances the efficiency of data retrieval in motif-based DNA storage systems.

cs.OH↗

arXiv2024

Coding Over Coupon Collector Channels for Combinatorial Motif-Based DNA Storage

Encoding information in combinations of pre-synthesised deoxyribonucleic acid (DNA) strands (referred to as motifs) is an interesting approach to DNA storage that could potentially circumvent the prohibitive costs of nucleotide-by-nucleotide DNA synthesis. Based on our analysis of an empirical data set from HelixWorks, we propose two channel models for this setup (with and without interference) and analyse their fundamental limits. We propose a coding scheme that approaches those limits by leveraging all information available at the output of the channel, in contrast to earlier schemes developed for a similar setup by Preuss et al. We highlight an important connection between channel capacity curves and the fundamental trade-off between synthesis (writing) and sequencing (reading), and offer a way to mitigate an exponential growth in decoding complexity with the size of the motif library.

cs.IT↗