An official website of the United States government
In high-performance computing (HPC), modern supercomputers typically provide exclusive computing resources to user applications. Nevertheless, the interconnect network is a shared resource for both inter-node communication and across-node I/O access, among co-running workloads, leading to inevitable network interference. In this study, we develop MFNetSim, a multi-fidelity modeling framework that enables simulation of multi-traffic simultaneously over the interconnect network, including inter-process communication and I/O traffic. By combining different levels of abstraction, MFNetSim can efficiently co-model the communication and I/O traffic occurring on HPC systems equipped with flash-based storage. We conduct simulation studies of hybrid workloads composed of traditional HPC applications and emerging ML applications on a 1,056-node Dragonfly system with various configurations. Our analysis provides various observations regarding how network interference affects communication and I/O traffic.
more »
« less
Prediction for distributional outcomes in high-performance computing input/output variability
Abstract Although high-performance computing (HPC) systems have been scaled to meet the exponentially growing demand for scientific computing, HPC performance variability remains a major challenge in computer science. Statistically, performance variability can be characterized by a distribution. Predicting performance variability is a critical step in HPC performance variability management. In this article, we propose a new framework to predict performance distributions. The proposed framework is a modified Gaussian process that can predict the distribution function of the input/output (I/O) throughput under a specific HPC system configuration. We also impose a monotonic constraint so that the predicted function is nondecreasing, which is a property of the cumulative distribution function. Additionally, the proposed model can incorporate both quantitative and qualitative input variables. We predict the HPC I/O distribution using the proposed method for the IOzone variability data. Data analysis results show that our framework can generate accurate predictions, and outperform existing methods. We also show how the predicted functional output can be used to generate predictions for a scalar summary of the performance distribution, such as the mean, standard deviation, and quantiles. Our prediction results can further be used for HPC system variability monitoring and optimization. This article has online supplementary materials.
more »
« less
- Award ID(s):
- 1838271
- PAR ID:
- 10487097
- Publisher / Repository:
- Oxford University Press
- Date Published:
- Journal Name:
- Journal of the Royal Statistical Society Series C: Applied Statistics
- Volume:
- 73
- Issue:
- 3
- ISSN:
- 0035-9254
- Format(s):
- Medium: X Size: p. 561-580
- Size(s):
- p. 561-580
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Deep Neural Networks (DNNs) have been applied as an effective machine learning algorithm to tackle problems in different domains. However, the endeavor to train sophisticated DNN models can stretch from days into weeks, presenting substantial obstacles in the realm of research focused on large-scale DNN architectures. Distributed Deep Learning (DDL) contributes to accelerating DNN training by distributing training workloads across multiple computation accelerators, for example, graphics processing units (GPUs). Despite the considerable amount of research directed toward enhancing DDL training, the influence of data loading on GPU utilization and overall training efficacy remains relatively overlooked. It is non-trivial to optimize data-loading in DDL applications that need intensive central processing unit (CPU) and input/output (I/O) resources to process enormous training data. When multiple DDL applications are deployed on a system (e.g., Cloud and High-Performance Computing (HPC) system), the lack of a practical and efficient technique for data-loader allocation incurs GPU idleness and degrades the training throughput. Therefore, our work first focuses on investigating the impact of data-loading on the global training throughput. We then propose a throughput prediction model to predict the maximum throughput for an individual DDL training application. By leveraging the predicted results, A-Dloader is designed to dynamically allocate CPU and I/O resources to concurrently running DDL applications and use the data-loader allocation as a knob to reduce GPU idle intervals and thus improve the overall training throughput. We implement and evaluate A-Dloader in a DDL framework for a series of DDL applications arriving and completing across the runtime. Our experimental results show that A-Dloader can achieve a 28.9% throughput improvement and a 10% makespan improvement compared with allocating resources evenly across applications.more » « less
-
null (Ed.)Parallel file systems (PFSes) and parallel I/O libraries have been the backbone of high-performance computing (HPC) infrastructures for decades. However, their crash consistency bugs have not been extensively studied, and the corresponding bug-finding or testing tools are lacking. In this paper, we first conduct a thorough bug study on the popular PFSes, such as BeeGFS and OrangeFS, with a cross-stack approach that covers HPC I/O library, PFS, and interactions with local file systems. The study results drive our design of a scalable testing framework, named PFSCHECK. PFSCHECK is easy to use with low performance overhead, as it can automatically generate test cases for triggering potential crash-consistency bugs, and trace essential file operations with low overhead. PFSCHECK is scalable for supporting large-scale HPC clusters, as it can exploit the parallelism to facilitate the verification of persistent storage states.more » « less
-
—Exascale computing enables unprecedented, detailed and coupled scientific simulations which generate data on the order of tens of petabytes. Due to large data volumes, lossy compressors become indispensable as they enable better compression ratios and runtime performance than lossless compressors. Moreover, as (high-performance computing) HPC systems grow larger, they draw power on the scale of tens of megawatts. Data motion is expensive in time and energy. Therefore, optimizing compressor and data I/O power usage is an important step in reducing energy consumption to meet sustainable computing goals and stay within limited power budgets. In this paper, we explore efficient power consumption gains for the SZ and ZFP lossy compressors and data writing on a cloud HPC system while varying the CPU frequency, scientific data sets, and system architecture. Using this power consumption data, we construct a power model for lossy compression and present a tuning methodology that reduces energy overhead of lossy compressors and data writing on HPC systems by 14.3% on average. We apply our model and find 6.5 kJs, or 13%, of savings on average for 512GB I/O. Therefore, utilizing our model results in more energy efficient lossy data compression and I/O.more » « less
-
Parallel File Systems (PFSs) are frequently deployed on leadership High Performance Computing (HPC) systems to ensure efficient I/O, persistent storage and scalable performance. Emerging Deep Learning (DL) applications incur new I/O and storage requirements to HPC systems with batched input of small random files. This mandates PFSs to have commensurate features that can meet the needs of DL applications. BeeGFS is a recently emerging PFS that has grabbed the attention of the research and industry world because of its performance, scalability and ease of use. While emphasizing a systematic performance analysis of BeeGFS, in this paper, we present the architectural and system features of BeeGFS, and perform an experimental evaluation using cutting-edge I/O, Metadata and DL application benchmarks. Particularly, we have utilized AlexNet and ResNet-50 models for the classification of ImageNet dataset using the Livermore Big Artificial Neural Network Toolkit (LBANN), and ImageNet data reader pipeline atop TensorFlow and Horovod. Through extensive performance characterization of BeeGFS, our study provides a useful documentation on how to leverage BeeGFS for the emerging DL applications.more » « less
