Searcharxiv⌕ Search

arXiv subjects

Franck Cappello

Publications and source records attributed to Franck Cappello.

84 records · Page 5Linked to original sources
arXiv2019

DeepSZ: A Novel Framework to Compress Deep Neural Networks by Using Error-Bounded Lossy Compression

DNNs have been quickly and broadly exploited to improve the data analysis quality in many complex science and engineering applications. Today's DNNs are becoming deeper and wider because of increasing demand on the analysis quality and more and more complex applications to resolve. The wide and deep DNNs, however, require large amounts of resources, significantly restricting their utilization on resource-constrained systems. Although some network simplification methods have been proposed to address this issue, they suffer from either low compression ratios or high compression errors, which may introduce a costly retraining process for the target accuracy. In this paper, we propose DeepSZ: an accuracy-loss bounded neural network compression framework, which involves four key steps: network pruning, error bound assessment, optimization for error bound configuration, and compressed model generation, featuring a high compression ratio and low encoding time. The contribution is three-fold. (1) We develop an adaptive approach to select the feasible error bounds for each layer. (2) We build a model to estimate the overall loss of accuracy based on the accuracy degradation caused by individual decompressed layers. (3) We develop an efficient optimization algorithm to determine the best-fit configuration of error bounds in order to maximize the compression ratio under the user-set accuracy constraint. Experiments show that DeepSZ can compress AlexNet and VGG-16 on the ImageNet by a compression ratio of 46X and 116X, respectively, and compress LeNet-300-100 and LeNet-5 on the MNIST by a compression ratio of 57X and 56X, respectively, with only up to 0.3% loss of accuracy. Compared with other state-of-the-art methods, DeepSZ can improve the compression ratio by up to 1.43X, the DNN encoding performance by up to 4.0X (with four Nvidia Tesla V100 GPUs), and the decoding performance by up to 6.2X.

cs.CV↗
arXiv2019

Optimizing Lossy Compression Rate-Distortion from Automatic Online Selection between SZ and ZFP

With ever-increasing volumes of scientific data produced by HPC applications, significantly reducing data size is critical because of limited capacity of storage space and potential bottlenecks on I/O or networks in writing/reading or transferring data. SZ and ZFP are the two leading lossy compressors available to compress scientific data sets. However, their performance is not consistent across different data sets and across different fields of some data sets: for some fields SZ provides better compression performance, while other fields are better compressed with ZFP. This situation raises the need for an automatic online (during compression) selection between SZ and ZFP, with a minimal overhead. In this paper, the automatic selection optimizes the rate-distortion, an important statistical quality metric based on the signal-to-noise ratio. To optimize for rate-distortion, we investigate the principles of SZ and ZFP. We then propose an efficient online, low-overhead selection algorithm that predicts the compression quality accurately for two compressors in early processing stages and selects the best-fit compressor for each data field. We implement the selection algorithm into an open-source library, and we evaluate the effectiveness of our proposed solution against plain SZ and ZFP in a parallel environment with 1,024 cores. Evaluation results on three data sets representing about 100 fields show that our selection algorithm improves the compression ratio up to 70% with the same level of data distortion because of very accurate selection (around 99%) of the best-fit compressor, with little overhead (less than 7% in the experiments).

cs.DC↗
arXiv2018

Amplitude-Aware Lossy Compression for Quantum Circuit Simulation

Classical simulation of quantum circuits is crucial for evaluating and validating the design of new quantum algorithms. However, the number of quantum state amplitudes increases exponentially with the number of qubits, leading to the exponential growth of the memory requirement for the simulations. In this paper, we present a new data reduction technique to reduce the memory requirement of quantum circuit simulations. We apply our amplitude-aware lossy compression technique to the quantum state amplitude vector to trade the computation time and fidelity for memory space. The experimental results show that our simulator only needs 1/16 of the original memory requirement to simulate Quantum Fourier Transform circuits with 99.95% fidelity. The reduction amount of memory requirement suggests that we could increase 4 qubits in the quantum circuit simulation comparing to the simulation without our technique. Additionally, for some specific circuits, like Grover's search, we could increase the simulation size by 18 qubits.

quant-ph↗
arXiv2018

Memory-Efficient Quantum Circuit Simulation by Using Lossy Data Compression

In order to evaluate, validate, and refine the design of new quantum algorithms or quantum computers, researchers and developers need methods to assess their correctness and fidelity. This requires the capabilities of quantum circuit simulations. However, the number of quantum state amplitudes increases exponentially with the number of qubits, leading to the exponential growth of the memory requirement for the simulations. In this work, we present our memory-efficient quantum circuit simulation by using lossy data compression. Our empirical data shows that we reduce the memory requirement to 16.5% and 2.24E-06 of the original requirement for QFT and Grover's search, respectively. This finding further suggests that we can simulate deep quantum circuits up to 63 qubits with 0.8 petabytes memory.

quant-ph↗
arXiv2018

Fixed-PSNR Lossy Compression for Scientific Data

Error-controlled lossy compression has been studied for years because of extremely large volumes of data being produced by today's scientific simulations. None of existing lossy compressors, however, allow users to fix the peak signal-to-noise ratio (PSNR) during compression, although PSNR has been considered as one of the most significant indicators to assess compression quality. In this paper, we propose a novel technique providing a fixed-PSNR lossy compression for scientific data sets. We implement our proposed method based on the SZ lossy compression framework and release the code as an open-source toolkit. We evaluate our fixed-PSNR compressor on three real-world high-performance computing data sets. Experiments show that our solution has a high accuracy in controlling PSNR, with an average deviation of 0.1 ~ 5.0 dB on the tested data sets.

cs.IT↗
arXiv2018

Improving Performance of Iterative Methods by Lossy Checkponting

Iterative methods are commonly used approaches to solve large, sparse linear systems, which are fundamental operations for many modern scientific simulations. When the large-scale iterative methods are running with a large number of ranks in parallel, they have to checkpoint the dynamic variables periodically in case of unavoidable fail-stop errors, requiring fast I/O systems and large storage space. To this end, significantly reducing the checkpointing overhead is critical to improving the overall performance of iterative methods. Our contribution is fourfold. (1) We propose a novel lossy checkpointing scheme that can significantly improve the checkpointing performance of iterative methods by leveraging lossy compressors. (2) We formulate a lossy checkpointing performance model and derive theoretically an upper bound for the extra number of iterations caused by the distortion of data in lossy checkpoints, in order to guarantee the performance improvement under the lossy checkpointing scheme. (3) We analyze the impact of lossy checkpointing (i.e., extra number of iterations caused by lossy checkpointing files) for multiple types of iterative methods. (4)We evaluate the lossy checkpointing scheme with optimal checkpointing intervals on a high-performance computing environment with 2,048 cores, using a well-known scientific computation package PETSc and a state-of-the-art checkpoint/restart toolkit. Experiments show that our optimized lossy checkpointing scheme can significantly reduce the fault tolerance overhead for iterative methods by 23%~70% compared with traditional checkpointing and 20%~58% compared with lossless-compressed checkpointing, in the presence of system failures.

cs.DC↗
arXiv2017

In-Depth Exploration of Single-Snapshot Lossy Compression Techniques for N-Body Simulations

In situ lossy compression allowing user-controlled data loss can significantly reduce the I/O burden. For large-scale N-body simulations where only one snapshot can be compressed at a time, the lossy compression ratio is very limited because of the fairly low spatial coherence of the particle data. In this work, we assess the state-of-the-art single-snapshot lossy compression techniques of two common N-body simulation models: cosmology and molecular dynamics. We design a series of novel optimization techniques based on the two representative real-world N-body simulation codes. For molecular dynamics simulation, we propose three compression modes (i.e., best speed, best tradeoff, best compression mode) that can refine the tradeoff between the compression rate (a.k.a., speed/throughput) and ratio. For cosmology simulation, we identify that our improved SZ is the best lossy compressor with respect to both compression ratio and rate. Its compression ratio is higher than the second-best compressor by 11% with comparable compression rate. Experiments with up to 1024 cores on the Blues supercomputer at Argonne show that our proposed lossy compression method can reduce I/O time by 80% compared with writing data directly to a parallel file system and outperforms the second-best solution by 60%. Moreover, our proposed lossy compression methods have the best rate-distortion with reasonable compression errors on the tested N-body simulation data compared with state-of-the-art compressors.

cs.DC↗
arXiv2017

Z-checker: A Framework for Assessing Lossy Compression of Scientific Data

Because of vast volume of data being produced by today's scientific simulations and experiments, lossy data compressor allowing user-controlled loss of accuracy during the compression is a relevant solution for significantly reducing the data size. However, lossy compressor developers and users are missing a tool to explore the features of scientific datasets and understand the data alteration after compression in a systematic and reliable way. To address this gap, we have designed and implemented a generic framework called Z-checker. On the one hand, Z-checker combines a battery of data analysis components for data compression. On the other hand, Z-checker is implemented as an open-source community tool to which users and developers can contribute and add new analysis components based on their additional analysis demands. In this paper, we present a survey of existing lossy compressors. Then we describe the design framework of Z-checker, in which we integrated evaluation metrics proposed in prior work as well as other analysis tools. Specifically, for lossy compressor developers, Z-checker can be used to characterize critical properties of any dataset to improve compression strategies. For lossy compression users, Z-checker can detect the compression quality, provide various global distortion analysis comparing the original data with the decompressed data and statistical analysis of the compression error. Z-checker can perform the analysis with either coarse granularity or fine granularity, such that the users and developers can select the best-fit, adaptive compressors for different parts of the dataset. Z-checker features a visualization interface displaying all analysis results in addition to some basic views of the datasets such as time series. To the best of our knowledge, Z-checker is the first tool designed to assess lossy compression comprehensively for scientific datasets.

cs.OH↗
arXiv2017

Exploration of Pattern-Matching Techniques for Lossy Compression on Cosmology Simulation Data Sets

Because of the vast volume of data being produced by today's scientific simulations, lossy compression allowing user-controlled information loss can significantly reduce the data size and the I/O burden. However, for large-scale cosmology simulation, such as the Hardware/Hybrid Accelerated Cosmology Code (HACC), where memory overhead constraints restrict compression to only one snapshot at a time, the lossy compression ratio is extremely limited because of the fairly low spatial coherence and high irregularity of the data. In this work, we propose a pattern-matching (similarity searching) technique to optimize the prediction accuracy and compression ratio of SZ lossy compressor on the HACC data sets. We evaluate our proposed method with different configurations and compare it with state-of-the-art lossy compressors. Experiments show that our proposed optimization approach can improve the prediction accuracy and reduce the compressed size of quantization codes compared with SZ. We present several lessons useful for future research involving pattern-matching techniques for lossy compression.

cs.IT↗
arXiv2017

Significantly Improving Lossy Compression for Scientific Data Sets Based on Multidimensional Prediction and Error-Controlled Quantization

Today's HPC applications are producing extremely large amounts of data, such that data storage and analysis are becoming more challenging for scientific research. In this work, we design a new error-controlled lossy compression algorithm for large-scale scientific data. Our key contribution is significantly improving the prediction hitting rate (or prediction accuracy) for each data point based on its nearby data values along multiple dimensions. We derive a series of multilayer prediction formulas and their unified formula in the context of data compression. One serious challenge is that the data prediction has to be performed based on the preceding decompressed values during the compression in order to guarantee the error bounds, which may degrade the prediction accuracy in turn. We explore the best layer for the prediction by considering the impact of compression errors on the prediction accuracy. Moreover, we propose an adaptive error-controlled quantization encoder, which can further improve the prediction hitting rate considerably. The data size can be reduced significantly after performing the variable-length encoding because of the uneven distribution produced by our quantization encoder. We evaluate the new compressor on production scientific data sets and compare it with many other state-of-the-art compressors: GZIP, FPZIP, ZFP, SZ-1.1, and ISABELA. Experiments show that our compressor is the best in class, especially with regard to compression factors (or bit-rates) and compression errors (including RMSE, NRMSE, and PSNR). Our solution is better than the second-best solution by more than a 2x increase in the compression factor and 3.8x reduction in the normalized root mean squared error on average, with reasonable error bounds and user-desired bit-rates.

cs.IT↗
arXiv2003

Augernome & XtremWeb: Monte Carlos computation on a global computing platform

In this paper, we present XtremWeb, a Global Computing platform used to generate monte carlos showers in Auger, an HEP experiment to study the highest energy cosmic rays at Mallargue-Mendoza, Argentina. XtremWeb main goal, as a Global Computing platform, is to compute distributed applications using idle time of widely interconnected machines. It is especially dedicated to -but not limited to- multi-parameters applications such as monte carlos computations; its security mechanisms ensuring not only hosts integrity but also results certification and its fault tolerant features, encouraged us to test it and, finally, to deploy it as to support our CPU needs to simulate showers. We first introduce Auger computing needs and how Global Computing could help. We then detail XtremWeb architecture and goals. The fourth and last part presents the profits we have gained to choose this platform. We conclude on what could be done next.

cs.DC↗