arXiv subjects

Daniel Waddington

Publications and source records attributed to Daniel Waddington.

7 recordsLinked to original sources

arXiv2026

RAGPerf: An End-to-End Benchmarking Framework for Retrieval-Augmented Generation Systems

We present the design and implementation of a RAG-based AI system benchmarking (RAGPerf) framework for characterizing the system behaviors of RAG pipelines. To facilitate detailed profiling and fine-grained performance analysis, RAGPerf decouples the RAG workflow into several modular components - embedding, indexing, retrieval, reranking, and generation. RAGPerf offers the flexibility for users to configure the core parameters of each component and examine their impact on the end-to-end query performance and quality. RAGPerf has a workload generator to model real-world scenarios by supporting diverse datasets (e.g., text, pdf, code, and audio), different retrieval and update ratios, and query distributions. RAGPerf also supports different embedding models, major vector databases such as LanceDB, Milvus, Qdrant, Chroma, and Elasticsearch, as well as different LLMs for content generation. It automates the collection of performance metrics (i.e., end-to-end query throughput, host/GPU memory footprint, and CPU/GPU utilization) and accuracy metrics (i.e., context recall, query accuracy, and factual consistency). We demonstrate the capabilities of RAGPerf through a comprehensive set of experiments and open source its codebase at GitHub. Our evaluation shows that RAGPerf incurs negligible performance overhead.

cs.PF↗

arXiv2025

Speeding up Model Loading with fastsafetensors

The rapid increases in model parameter sizes introduces new challenges in pre-trained model loading. Currently, machine learning code often deserializes each parameter as a tensor object in host memory before copying it to device memory. We found that this approach underutilized storage throughput and significantly slowed down loading large models with a widely-used model file formats, safetensors. In this work, we present fastsafetensors, a Python library designed to optimize the deserialization of tensors in safetensors files. Our approach first copies groups of on-disk parameters to device memory, where they are directly instantiated as tensor objects. This design enables further optimization in low-level I/O and high-level tensor preprocessing, including parallelized copying, peer-to-peer DMA, and GPU offloading. Experimental results show performance improvements of 4.8x to 7.5x in loading models such as Llama (7, 13, and 70 billion parameters), Falcon (40 billion parameters), and the Bloom (176 billion parameters).

cs.DC↗

arXiv2025

Lossless Compression for LLM Tensor Incremental Snapshots

During the training of Large Language Models (LLMs), tensor data is periodically "checkpointed" to persistent storage to allow recovery of work done in the event of failure. The volume of data that must be copied during each checkpoint, even when using reduced-precision representations such as bfloat16, often reaches hundreds of gigabytes. Furthermore, the data must be moved across a network and written to a storage system before the next epoch occurs. With a view to ultimately building an optimized checkpointing solution, this paper presents experimental analysis of checkpoint data used to derive a design that maximizes the use of lossless compression to reduce the volume of data. We examine how tensor data and its compressibility evolve during model training and evaluate the efficacy of existing common off-the-shelf general purpose compression engines combined with known data optimization techniques such as byte-grouping and incremental delta compression. Leveraging our analysis we have built an effective compression solution, known as Language Model Compressor (LMC), which is based on byte-grouping and Huffman encoding. LMC offers more compression performance than the best alternative (BZ2) but with an order-of-magnitude reduction in the time needed to perform the compression. We show that a 16-core parallel implementation of LMC can attain compression and decompression throughput of 2.78 GiB/s and 3.76 GiB/s respectively. This increase in performance ultimately reduces the CPU resources needed and provides more time to copy the data to the storage system before the next epoch thus allowing for higher-frequency checkpoints.

cs.LG↗

arXiv2022

Non-Volatile Memory Accelerated Posterior Estimation

Bayesian inference allows machine learning models to express uncertainty. Current machine learning models use only a single learnable parameter combination when making predictions, and as a result are highly overconfident when their predictions are wrong. To use more learnable parameter combinations efficiently, these samples must be drawn from the posterior distribution. Unfortunately computing the posterior directly is infeasible, so often researchers approximate it with a well known distribution such as a Gaussian. In this paper, we show that through the use of high-capacity persistent storage, models whose posterior distribution was too big to approximate are now feasible, leading to improved predictions in downstream tasks.

cs.LG↗

arXiv2022

Non-Volatile Memory Accelerated Geometric Multi-Scale Resolution Analysis

Dimensionality reduction algorithms are standard tools in a researcher's toolbox. Dimensionality reduction algorithms are frequently used to augment downstream tasks such as machine learning, data science, and also are exploratory methods for understanding complex phenomena. For instance, dimensionality reduction is commonly used in Biology as well as Neuroscience to understand data collected from biological subjects. However, dimensionality reduction techniques are limited by the von-Neumann architectures that they execute on. Specifically, data intensive algorithms such as dimensionality reduction techniques often require fast, high capacity, persistent memory which historically hardware has been unable to provide at the same time. In this paper, we present a re-implementation of an existing dimensionality reduction technique called Geometric Multi-Scale Resolution Analysis (GMRA) which has been accelerated via novel persistent memory technology called Memory Centric Active Storage (MCAS). Our implementation uses a specialized version of MCAS called PyMM that provides native support for Python datatypes including NumPy arrays and PyTorch tensors. We compare our PyMM implementation against a DRAM implementation, and show that when data fits in DRAM, PyMM offers competitive runtimes. When data does not fit in DRAM, our PyMM implementation is still able to process the data.

cs.LG↗

arXiv2021

An Architecture for Memory Centric Active Storage (MCAS)

The advent of CPU-attached persistent memory technology, such as Intel's Optane Persistent Memory Modules (PMM), has brought with it new opportunities for storage. In 2018, IBM Research Almaden began investigating and developing a new enterprise-grade storage solution directly aimed at this emerging technology. MCAS (Memory Centric Active Storage) defines an evolved network-attached key-value store that offers both near-data compute and the ability to layer enterprise-grade data management services on shared persistent memory. As a converged memory-storage tier, MCAS moves towards eliminating the traditional separation of compute and storage, and thereby unifying the data space. This paper provides an in-depth review of the MCAS architecture and implementation, as well as general performance results.

cs.AR↗

arXiv2021

A High-Performance Persistent Memory Key-Value Store with Near-Memory Compute

MCAS (Memory Centric Active Storage) is a persistent memory tier for high-performance durable data storage. It is designed from the ground-up to provide a key-value capability with low-latency guarantees and data durability through memory persistence and replication. To reduce data movement and make further gains in performance, we provide support for user-defined "push-down" operations (known as Active Data Objects) that can execute directly and safely on the value-memory associated with one or more keys. The ADO mechanism allows complex pointer-based dynamic data structures (e.g., trees) to be stored and operated on in persistent memory. To this end, we examine a real-world use case for MCAS-ADO in the handling of enterprise storage system metadata for Continuous Data Protection (CDP). This requires continuously updating complex metadata that must be kept consistent and durable. In this paper, we i.) present the MCAS-ADO system architecture, ii.) show how the CDP use case is implemented, and finally iii.) give an evaluation of system performance in the context of this use case.

cs.DB↗