Search NASA⌕ Search

Engineering topics

Ogrenci-Memik, Seda

Publications and source records attributed to Ogrenci-Memik, Seda.

Applications and Techniques for Fast Machine Learning in Science

In this community review report, we discuss applications and techniques for fast machine learning (ML) in science—the concept of integrating powerful ML methods into the real-time experimental data processing loop to accelerate scientific discovery. The material for the report builds on two workshops held by the Fast ML for Science community and covers three main areas: applications for fast ML across a number of scientific domains; techniques for training and implementing performant and resource-efficient ML algorithms; and computing architectures, platforms, and technologies for deploying these algorithms. We also present overlapping challenges across the multiple scientific domains where common solutions can be found. This community report is intended to give plenty of examples and inspiration for scientific discovery through integrated and accelerated ML solutions. This is followed by a high-level overview and organization of technical advances, including an abundance of pointers to source material, which can enable these breakthroughs.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

A Reconfigurable Neural Network ASIC for Detector Front-End Data Compression at the HL-LHC

Despite advances in the programmable logic capabilities of modern trigger systems, a significant bottleneck remains in the amount of data to be transported from the detector to off-detector logic where trigger decisions are made. We demonstrate that a neural network (NN) autoencoder model can be implemented in a radiation-tolerant application-specific integrated circuit (ASIC) to perform lossy data compression alleviating the data transmission problem while preserving critical information of the detector energy profile. For our application, we consider the high-granularity calorimeter from the Compact Muon Solenoid (CMS) experiment at the CERN Large Hadron Collider. The advantage of the machine learning approach is in the flexibility and configurability of the algorithm. By changing the NN weights, a unique data compression algorithm can be deployed for each sensor in different detector regions and changing detector or collider conditions. To meet area, performance, and power constraints, we perform quantization-aware training to create an optimized NN hardware implementation. The design is achieved through the use of high-level synthesis tools and the hls4ml framework and was processed through synthesis and physical layout flows based on a low-power (LP)-CMOS 65-nm technology node. The flow anticipates 200 Mrad of ionizing radiation to select gates and reports a total area of 3.6 mm 2 and consumes 95 mW of power. The simulated energy consumption per inference is 2.4 nJ. Furthermore, this is the first radiation-tolerant on-detector ASIC implementation of an NN that has been designed for particle physics applications.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Thermal Management for FPGA Nodes in HPC Systems

The integration of FPGAs into large-scale computing systems is gaining attention. In these systems, real-time data handling for networking, tasks for scientific computing, and machine learning can be executed with customized datapaths on reconfigurable fabric within heterogeneous compute nodes. At the same time, thermal management, particularly battling the cooling cost and guaranteeing the reliability, is a continuing concern. The introduction of new heterogeneous components into HPC nodes only adds further complexities to thermal modeling and management. The thermal behavior of multi-FPGA systems deployed within large compute clusters is less explored. Here, we first show that the thermal behaviors of different FPGAs of the same generation can vary due to their physical locations in a rack and process variation, even though they are running the same tasks. We present a machine learning–based model to capture the thermal behavior of each individual FPGA in the cluster. We then propose two thermal management strategies guided by our thermal model. First, we mitigate thermal variation and hotspots across the cluster by proactive thermal-aware task placement. Under the tested system and benchmarks, we achieve up to 26.4° C and on average 13.3° C system temperature reduction with no performance penalty. Second, we utilize this thermal model to guide HLS parameter tuning at the task design stage to achieve improved thermal response after deployment.

97 MATHEMATICS AND COMPUTING↗

The ETROC Project: Precision Timing ASIC Development for LGAD-based CMS Endcap Timing Layer (ETL) Upgrade

The ETROC (Endcap Timing Readout Chip) is being developed for the LGAD-based CMS Endcap Timing Layer (ETL) at HL-LHC. The ETL on each side of the interaction region will be instrumented with a two-disk system of MIP-sensitive LGAD silicon devices to be read out by ETROCs for precision timing measurement with down to 30 ps timing resolution. The ETROC is designed to handle a 16 x 16 pixel cell matrix, with each pixel cell being 1.3 mm x 1.3 mm to match the LGAD sensor pixel size. Approximately 15% of the sensors near the highest eta region will experience hadron fluence above 1e15 neq/cm² towards end of operation of HL-LHC, resulting in small signal amplitude with LGAD gain reduced to around 10. For this reason, the front-end design for preamplifier and discriminator has been specifically optimized for the reduced LGAD signals, with enough flexibilities to meet the ETL specific needs for time resolution, power budget and radiation profile.

97 MATHEMATICS AND COMPUTING↗

A Low Power, High Speed Readout for Pixel Detectors Based on an Arbitration Tree

A low-power, high-speed arbitration tree for pixel detector readout is presented. The synchronized, binary tree priority encoder establishes a position dependent priority list at the start of every time frame. Pixels that indicate the presence of data for readout, are sequentially granted access to a shared bus for data transfer to the periphery, without the use of an additional global strobe signal. It can be used for either full frame imaging or zero-suppressed readout, in which case it can simultaneously generate the pixel address. To increase the readout frame rate, the pixel array is subdivided into two halves, which allows interleaved latching of data at the output serializer. The design was implemented in a 65 nm LP-CMOS process for the readout of a 64 x 64 pixel array. Measurement results demonstrate a deadtime-less, full frame imaging rate of ~50 kfps, achieved with a dedicated output for every (32 x 32) 1024 pixels and for a pixel data packet of 11 bits, with no bit errors detected over 1000 frames. The measured energy per bit is 0.94 pJ.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Performance Study of the First 2D Prototype of Vertically Integrated Pattern Recognition Associative Memory (VIPRAM)

Extremely fast pattern recognition capabilities are necessary to find and fit billions of tracks at the hardware trigger level produced every second anticipated at high luminosity LHC (HL-LHC) running conditions. Associative Memory (AM) based approaches for fast pattern recognition have been proposed as a potential solution to the tracking trigger. However, at the HL-LHC, there is much less time available and speed performance must be improved over previous systems while maintaining a comparable number of patterns. The Vertically Integrated Pattern Recognition Associative Memory (VIPRAM) Project aims to achieve the target pattern density and performance goal using 3DIC technology. The first step taken in the VIPRAM work was the development of a 2D prototype (protoVIPRAM00) in which the associative memory building blocks were designed to be compatible with the 3D integration. In this paper, we present the results from extensive performance studies of the protoVIPRAM00 chip in both realistic HL-LHC and extreme conditions. Results indicate that the chip operates at the design frequency of 100 MHz with perfect correctness in realistic conditions and conclude that the building blocks are ready for 3D stacking. We also present performance boundary characterization of the chip under extreme conditions.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Machine Learning-Based Temperature Prediction for Runtime Thermal Management Across System Components

Elevated temperatures limit the peak performance of systems because of frequent interventions by thermal throttling. Non-uniform thermal states across system nodes also cause performance variation within seemingly equivalent nodes leading to significant degradation of overall performance. In this paper we present a framework for creating a lightweight thermal prediction system suitable for run-time management decisions. We pursue two avenues to explore optimized lightweight thermal predictors. First, we use feature selection algorithms to improve the performance of previously designed machine learning methods. Second, we develop alternative methods using neural network and linear regression-based methods to perform a comprehensive comparative study of prediction methods. We show that our optimized models achieve improved performance with better prediction accuracy and lower overhead as compared with the Gaussian process model proposed previously. Specifically we present a reduced version of the Gaussian process model, a neural network-based model, and a linear regression-based model. Using the optimization methods, we are able to reduce the average prediction errors in the Gaussian process from 4: 2 degrees C to 2: 9 degrees C. We also show that the newly developed models using neural network and Lasso linear regression have average prediction errors of 2: 9 degrees C and 3: 8 degrees C respectively. The prediction overheads are 0.22, 0.097, and 0.026 ms per prediction for reduced Gaussian process, neural network, and Lasso linear regression models, respectively, compared with 0.57 ms per prediction for the previous Gaussian process model. We have implemented our proposed thermal prediction models on a two-node system configuration to help identify the optimal task placement. The task placement identified by the models reduces the average system temperature by up to 11: 9 degrees C without any performance degradation. Furthermore, these models respectively achieve 75, 82.5, and 74.17 percent success rates in correctly pointing to those task placements with better thermal response, compared with 72.5 percent success for the original model in achieving the same objective. Lastly, we extended our analysis to a 16-node system and we were able to train models and execute them in real time to guide task migration and achieve on average 17 percent reduction in the overall system cooling power.

97 MATHEMATICS AND COMPUTING↗

Addressing Thermal and Performance Variability Issues in Dynamic Processors

As CMOS scaling nears its end, parameter variations (process, temperature and voltage) are becoming a major concern. To overcome parameter variations and provide stability, modern processors are becoming dynamic, opportunistically adjusting voltage and frequency based on thermal and energy constraints, which negatively impacts traditional bulk-synchronous parallelism-minded hardware and software designs. As node-level architecture is growing in complexity, implementing variation control mechanisms only with hardware can be a challenging task. In this paper we investigate a software strategy to manage hardwareinduced variations, leveraging low-level monitoring/controlling mechanisms.

97 MATHEMATICS AND COMPUTING↗