Related papers: Optimizing STAR Aligner for High Throughput Comput…
Significant progress has been made recently in point cloud segmentation utilizing an encoder-decoder framework, which initially encodes point clouds into low-resolution representations and subsequently decodes high-resolution predictions.…
Pipelining between data loading and computation is a critical tensor program optimization for GPUs. In order to unleash the high performance of latest GPUs, we must perform a synergetic optimization of multi-stage pipelining across the…
As comprehensive large model evaluation becomes prohibitively expensive, predicting model performance from limited observations has become essential. However, existing statistical methods struggle with pattern shifts, data sparsity, and…
This paper presents a stream-oriented architecture for structuring cluster applications. Clusters that run applications based on this architecture can scale to tenths of thousands of nodes with significantly less performance loss or…
We present the results of our investigations into options for the computing platform for the imaging pipeline in the CHILES project, an ultra-deep HI pathfinder for the era of the Square Kilometre Array. CHILES pushes the current computing…
Modern astronomical data processing requires complex software pipelines to process ever growing datasets. For radio astronomy, these pipelines have become so large that they need to be distributed across a computational cluster. This makes…
We propose the Chunkwise Aligner, a novel architecture for streaming automatic speech recognition (ASR). While the Transducer is the standard model for streaming ASR, its training is costly due to the need to compute all possible…
This paper presents a methodology to efficiently explore the design space of communication adapters. In most digital signal processing (DSP) applications, the overall performance of the system is significantly affected by communication…
Radio astronomy relies heavily on efficient and accurate processing pipelines to deliver science ready data. With the increasing data flow of modern radio telescopes, manual configuration of such data processing pipelines is infeasible.…
This paper explores a prevailing trend in the industry: migrating data-intensive analytics applications from on-premises to cloud-native environments. We find that the unique cost models associated with cloud-based storage necessitate a…
Data scientists seeking a good supervised learning model on a new dataset have many choices to make: they must preprocess the data, select features, possibly reduce the dimension, select an estimation algorithm, and choose hyperparameters…
Deep convolutional neural networks (CNN) are widely used in modern artificial intelligence (AI) and smart vision systems but also limited by computation latency, throughput, and energy efficiency on a resource-limited scenario, such as…
The volume of data generated by modern astronomical telescopes is extremely large and rapidly growing. However, current high-performance data processing architectures/frameworks are not well suited for astronomers because of their…
The field of connectomics faces unprecedented "big data" challenges. To reconstruct neuronal connectivity, automated pixel-level segmentation is required for petabytes of streaming electron microscopy data. Existing algorithms provide…
Fully re-orientable small spacecraft are now supported by commercial technologies, allowing them to point their instruments in any direction and capture images, with short notice. When combined with improved onboard processing, and…
The challenge to fully exploit the potential of existing and upcoming scientific instruments like large single-dish radio telescopes is to process the collected massive data effectively and efficiently. As a "quasi 2D stencil computation"…
Cloud computing is a style of computing in which dynamically scalable and other virtualized resources are provided as a service over the Internet. The energy consumption and makespan associated with the resources allocated should be taken…
The rapid evolution of artificial intelligence (AI) is leading to a new generation of hardware accelerators optimized for deep learning. Some of the designs of these accelerators are general enough to allow their use for other…
The rapid adoption of AI-powered applications demands high-performance, scalable, and efficient cloud database solutions, as traditional architectures often struggle with AI-driven workloads requiring real-time data access, vector search,…
Over the past two decades, helioseismology has revolutionized our understanding of the interior structure and dynamics of the Sun. Asteroseismology will soon place this knowledge into a broader context by providing structural data for…