Engineering Papers⌕ Search

Engineering topics

Li, Ang

Publications and source records attributed to Li, Ang.

139 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Detecting Anomalous Computation with RNNs on GPU-Accelerated HPC Machines

This paper presents a workload classification framework that discriminates illicit computation from authorized workloads on GPU-accelerated HPC systems. As such systems become more and more powerful, they are exploited by attackers to run malicious and for-profit programs that typically require extremely high computing ability to be successful. Our classification framework leverages the distinctive signatures between illicit and authorized workloads, and explore machine learning methods to learn the workloads and classify them. The framework uses lightweight, non-intrusive workload profiling to collect model input data, and explores multiple machine learning methods, particularly recurrent neural network (RNN) that is suitable for online anomalous workload detection. Evaluation results on three generations of GPU machines demonstrate that the workload classification framework can tell apart the illicit authorized workloads with a high accuracy of over 95%.

Pengfei, Zou↗
DOE OSTI2020

Orbital selectivity of layer-resolved tunneling in the iron-based superconductor Ba 0.6 K 0.4 Fe 2 As 2

Here, we use scanning tunneling microscopy/spectroscopy to elucidate the Cooper pairing of the iron pnictide superconductor Ba 0.6 K 0.4 Fe 2 As 2 . By a cold-cleaving technique, we obtain atomically resolved termination surfaces with different layer identities. Remarkably, we observe that the low-energy tunneling spectrum related to superconductivity has an unprecedented dependence on the layer identity. By cross referencing with the angle-revolved photoemission results and the tunneling data of LiFeAs, we find that tunneling on each termination surface probes superconductivity through selecting distinct Fe-3$\textit{d}$ orbitals. These findings imply the real-space orbital features of the Cooper pairing in the iron pnictide superconductors, and propose a general concept that, for complex multiorbital material, tunneling on different terminating layers can feature orbital selectivity.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2020

Synthesis of Interface-Driven Tunable Bandgap Metal Oxides

Mixed bandgap and bandgap tunability in semiconductors is critical in expanding their use. Composition altera-tions through single-crystal epitaxial growth and formation of multi-layer tandem structures are often employed to achieve mixed bandgaps albeit with limited tunability. In this study, self-assembled one-dimensional coordination polymers provide a fac-ile synthon and template for graphitic C-doped mesoporous oxides, gC-β-Ga 2 O 3 or gC-In 2 O 3 via controlled oxidative ligand ablation. These materials have mixed bandgaps, and colors, depending on amount of gC present. The carbon-oxide interface leads to induced gap states, hence, a stoichiometrically tunable band structure. Structurally, a multi-scale porous network percolating throughout the material is realized. The nature of heat-treatment and top-down process allows for facile tunabil-ity and formation of mixed bandgap metal oxides through controlled carbon deposition. As a proof of concept, gC-β-Ga 2 O 3 was utilized as a photocatalyst for CO 2 reduction, which demonstrated excellent selectivity and conversion rates into CH 4 and CO.

36 MATERIALS SCIENCE↗
DOE OSTI2020

FPDeep: Scalable Acceleration of CNN Training on Deeply-Pipelined FPGA Clusters

In this paper, we propose a framework called FPDeep, which uses a hybrid of model and layer paral- lelism to configure distributed reconfigurable clusters to train DNNs. This approach has numerous benefits. First, the design does not suffer from batch size growth. Second, novel workload and weight partitioning leads to balanced loads of both among nodes. And third, the entire system is fine-grained pipeline. This leads to high parallelism and utilization and also minimizes the time features need to be cached while waiting for back-propagation.

Wang, Tianqi↗
DOE OSTI2020

pnnl/TCBNN

New bit capability for GPUs, bit-tensor-core.

Li, Ang↗
DOE OSTI2020

CSB-RNN: A Faster-Than-Realtime RNN Acceleration Framework with Compressed Structured Blocks

Recurrent Neural Networks (RNN) is widely applied to temporal sequence analysis, where real-time performance is usually in demand. However, RNN suffers a heavy computational workload as the model comes with a large weight matrix. To alleviate the pain, model compression (pruning) schemes have been proposed for RNN that pruning the redundant (near-zero) weight-values. On the one hand, the non-structured pruning methods achieve a considerable pruning rate while bringing the computational irregularity, which is un-friendly to parallel-hardware. On the other hand, the existing structured pruning methods consider the hardware parallelism; However, they suffer a poor pruning rate due to the restrict constraints on pruning structure. This paper presents CSB-RNN, an optimized full-stack RNN framework with the novel compressed structured block (CSB) technique. The CSB-pruned RNN model comes with both fine-granularity that benefits the pruning rate and regular structure that facilitates the hardware-parallelism. Further, we propose a novel hardware architecture for inferencing the CSB-pruned model. Different from conventional parallel hardware, this architecture solves the block-workload imbalance issue and achieves an over 95% hardware utilization. With the experiments on 10 RNN models in 5 application domains, the CSB-RNN realizes 7×-20× lossless compression and up to 50× acceptable lossy-compression, which is 2×-7× to the prior art. With the addition of the novel hardware, the compressed-RNN inference reaches a super real-time latency of 10-400µs with FPGA implementation.

Shi, Runbin↗
DOE OSTI2020

Engineering unsymmetrically coordinated Cu-S 1 N 3 single atom sites with enhanced oxygen reduction activity

Atomic interface regulation is thought to be an efficient method to adjust the performance of single atom catalysts. Herein, a practical strategy was reported to rationally design single copper atoms coordinated with both sulfur and nitrogen atoms in metal-organic framework derived hierarchically porous carbon (S-Cu-ISA/SNC). The atomic interface configuration of the copper site in S-Cu-ISA/SNC is detected to be an unsymmetrically arranged Cu-S 1 N 3 moiety. The catalyst exhibits excellent oxygen reduction reaction activity with a half-wave potential of 0.918 V vs. RHE. Additionally, through in situ X-ray absorption fine structure tests, we discover that the low-valent Cuprous-S 1 N 3 moiety acts as an active center during the oxygen reduction process. Our discovery provides a universal scheme for the controllable synthesis and performance regulation of single metal atom catalysts toward energy applications.

25 ENERGY STORAGE↗
DOE OSTI2020

pnnl/QASMBench

A low-level OpenQASM benchmark suite for NISQ evaluation and simulation. Please see our paper for details. n this repository you will find a low-level and light-weighted benchmark suite based on IBM OpenQASM language (see spec). It collects commonly seen quantum algorithms and routines from various domains including chemistry, simulation, linear algebra, searching, optimization, arithmetic, machine learning, fault tolerance, cryptography, etc. QASMBench trades-off between generality and usability, covering the number of qubits ranging from 2 to 60K, and the circuit depth from 4 to 12M

Li, Ang↗
DOE OSTI2020

Indicator-directed Dynamic Power Management for Iterative Workloads on GPU-Accelerated Systems

Modern high-performance and warehouse computing centers show strong interest in minimizing system power consumption while satisfying customers’ quality of service (QoS). Dynamic voltage and frequency scaling (DVFS) is effective for achieving this goal. Nevertheless, automating the process online and making it transparent to users must address three major challenges: (1) Complexity — today’s hardware components (e.g., CPUs, GPUs, memory, network, etc.) can be configured in several or dozens of frequency/voltage states for satisfying divergent system demands. Given their combination and the emergence of heterogeneity, searching the optimal configuration in the design space online can be timing consuming. (2) QoS guarantee — user-defined objectives such as power constraint and performance target must be monitored, predicted and ensured at the best effort. (3) Adaptability — various known and unknown workloads run on systems. Workloads characteristics should be quickly determined and configurations dynamically adjusted in accord with workloads and QoS. In this work, we focus on applications exhibiting an interesting feature – iterative or periodic, which is common among conventional HPC and emerging machine learning workloads. We propose an online dynamic power-performance (ODPP) management framework to dynamically adjust GPU DVFS configurations to meet performance and power objectives and constraints, without any code annotation or intrusion. Particularly, ODPP extracts the performance and power indicators for applications from their resources utilization profiles in a short episode. It further automatically constructs an accurate model that infers from the indicators how the application's performance and power vary with GPU core and memory frequencies. Aided with the model, for both seen and unseen applications, ODPP can quickly determine the most appropriate DVFS configuration for their execution. We evaluate ODPP on an NVIDIA GPU using multiple exascale computing (ECP) and deep learning applications.

Zou, Pengfei↗
DOE OSTI2020

s-blas

This package includes the implementation for four sparse linear algebra kernels: Sparse-Matrix-Vector-Multiplication (SpMV), Sparse-Triangular-Solve (SpTRSV), Sparse-Matrix-Transposition (SpTrans) and Sparse-Matrix-Matrix-Multiplication (SpMM) for Single-node Multi-GPU (scale-up) platforms such as NVIDIA DGX-1 and DGX-2.

Li, Ang↗
DOE OSTI2020

Evaluating Modern GPU Interconnect: PCIe, NVLink, NV-SLI, NVSwitch and GPUDirect

High performance multi-GPU computing becomes an inevitable trend due to the ever-increasing demand on computation capability in emerging domains such as deep learning, big data and planet-scale simulations. However, the lack of deep understanding on how modern GPUs can be connected and the real impact of state-of-the-art interconnect technology on multi-GPU application performance become a hurdle. In this paper, we fill the gap by conducting a thorough evaluation on five latest types of modern GPU interconnects: PCIe, NVLink-V1, NVLink-V2, NVLink-SLI and NVSwitch, from six high-end servers and HPC platforms: NVIDIA P100-DGX-1, V100-DGX-1, DGX-2, OLCF’s SummitDev and Summit supercomputers, as well as an SLI-linked system with two NVIDIA Turing RTX-2080 GPUs. Based on the empirical evaluation, we have observed four new types of GPU communication network NUMA effects: three are triggered by NVLink’s topology, connectivity and routing, while one is caused by PCIe chipset design issue. These observations indicate that, for a multi-GPU application running in a multi-GPU node, choosing the right GPU combination can impose considerable impact on GPU communication efficiency, as well as an application’s overall performance. Our evaluation can be leveraged in building practical multi-GPU performance models, which are vital for GPU task allocation, scheduling and migration in a shared environment (e.g., AI cloud and HPC centers), as well as communication-oriented performance tuning.

Performance Evaluation, GPU, Interconnect, NUMA, P↗
DOE OSTI2019

Fingerprinting Anomalous Computation with RNN for GPU-accelerated HPC Machines

This paper presents a workload classification framework that discriminates illicit computation from authorized workloads on GPU-accelerated HPC systems. As such systems become more and more powerful, they are exploited by attackers to run malicious and for-profit programs that typically require extremely high computing ability to be successful. Our classification framework leverages the distinctive signatures between illicit and authorized workloads, and explore machine learning methods to learn the workloads and classify them. The framework uses lightweight, non-intrusive workload profiling to collect model input data, and explores multiple machine learning methods, particularly recurrent neural network (RNN) that is suitable for online anomalous workload detection. Evaluation results on four generations of GPU machines demonstrate that the workload classification framework can tell apart the illicit authorized workloads with a high accuracy of over 95%.

Zou, Pengfei↗