Gennady Pekhimenko

发表

SysML: The New Frontier of Machine Learning Systems pdf

Gustavo Alonso, Dan Alistarh, Dimitris S. Papailiopoulos, 2019, ArXiv.

RowClone: Fast and energy-efficient in-DRAM bulk data copy and initialization

Rachata Ausavarungnirun, Chris Fallin, Onur Mutlu, 2013, 2013 46th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).

Base-Delta-Immediate Compression: A Practical Data Compression Mechanism for On-Chip Caches

Onur Mutlu, Todd C. Mowry, Gennady Pekhimenko, 2012 .

Adaptive-latency DRAM: Optimizing DRAM timing for the common-case

Onur Mutlu, Gennady Pekhimenko, Vivek Seshadri, 2015, 2015 IEEE 21st International Symposium on High Performance Computer Architecture (HPCA).

Zorua: A holistic approach to resource virtualization in GPUs

Onur Mutlu, Samira Manabi Khan, Gennady Pekhimenko, 2016, 2016 49th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).

Exploiting compressed block size as an indicator of future reuse

Onur Mutlu, Todd C. Mowry, Gennady Pekhimenko, 2015, 2015 IEEE 21st International Symposium on High Performance Computer Architecture (HPCA).

TBD: Benchmarking and Analyzing Deep Neural Network Training pdf

Amar Phanishayee, Bianca Schroeder, Gennady Pekhimenko, 2018, ArXiv.

MLPerf Training Benchmark pdf

David Patterson, Carole-Jean Wu, Gu-Yeon Wei, 2019, MLSys.

Page overlays: An enhanced virtual memory framework to enable fine-grained memory management

Onur Mutlu, Olatunji Ruwase, Todd C. Mowry, 2015, 2015 ACM/IEEE 42nd Annual International Symposium on Computer Architecture (ISCA).

Base-delta-immediate compression: Practical data compression for on-chip caches

Onur Mutlu, Todd C. Mowry, Gennady Pekhimenko, 2012, 2012 21st International Conference on Parallel Architectures and Compilation Techniques (PACT).

Adaptive-Latency DRAM: Reducing DRAM Latency by Exploiting Timing Margins pdf

Onur Mutlu, Samira Manabi Khan, Gennady Pekhimenko, 2018, ArXiv.

Toggle-Aware Bandwidth Compression for GPUs

Gennady Pekhimenko, O. Mutlu, T. Mowry, 2015 .

ChargeCache: Reducing DRAM latency by exploiting row access locality

Onur Mutlu, Oguz Ergin, Gennady Pekhimenko, 2016, 2016 IEEE International Symposium on High Performance Computer Architecture (HPCA).

Linearly compressed pages: A low-complexity, low-latency main memory compression framework

Onur Mutlu, Todd C. Mowry, Gennady Pekhimenko, 2013, 2013 46th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).

RowClone: Accelerating Data Movement and Initialization Using DRAM pdf

Rachata Ausavarungnirun, Chris Fallin, Onur Mutlu, 2018, ArXiv.

RowClone: Fast and Efficient In-DRAM Copy and Initialization of Bulk Data

Rachata Ausavarungnirun, Chris Fallin, Onur Mutlu, 2013 .

Exploiting Row-Level Temporal Locality in DRAM to Reduce the Memory Access Latency pdf

Onur Mutlu, Oguz Ergin, Gennady Pekhimenko, 2018, ArXiv.

MLSys: The New Frontier of Machine Learning Systems

Christopher De Sa, Michael I. Jordan, Phillip B. Gibbons, 2019, 1904.03257.

Rollback-free value prediction with approximate loads

Onur Mutlu, Todd C. Mowry, Gennady Pekhimenko, 2014, 2014 23rd International Conference on Parallel Architecture and Compilation (PACT).

Linearly compressed pages: A main memory compression framework with low complexity and low latency

Onur Mutlu, Todd C. Mowry, Gennady Pekhimenko, 2012, 2012 21st International Conference on Parallel Architectures and Compilation Techniques (PACT).

Simultaneous Multi Layer Access: A High Bandwidth and Low Cost 3D-Stacked Memory Interface pdf

Onur Mutlu, Samira Manabi Khan, Gennady Pekhimenko, 2015, ArXiv.

Simultaneous Multi-Layer Access

Onur Mutlu, Samira Manabi Khan, Gennady Pekhimenko, 2016, ACM Trans. Archit. Code Optim..

Approximate Computing : Application Analysis and Hardware Design

Gennady Pekhimenko, Danai Koutra, Kun Qian, 2011 .

RFVP: Rollback-Free Value Prediction with Safe-to-Approximate Loads

Onur Mutlu, Todd C. Mowry, Gennady Pekhimenko, 2016, ACM Trans. Archit. Code Optim..

DNN-Train : Benchmarking and Analyzing DNN Training

Gennady Pekhimenko, Hongyu Zhu, Gennady Pekhimenko, 2018 .

Machine Learning Algorithms for Choosing Compiler Heuristics Abstract Machine Learning Algorithms for Choosing Compiler Heuristics

Gennady Pekhimenko, 2008 .

Java Virtual Machine Interpreter : Porting and Extending . Final Report CSC 2227

Gennady Pekhimenko, 2007 .

How to validate Machine Learning Models Prior to Deployment: Silent trial protocol for evaluation of real-time models at ICU

Gennady Pekhimenko, S. Tonekaboni, M. McCradden, 2022, CHIL.

DietCode: Automatic Optimization for Dynamic Tensor Programs

Cody Hao Yu, Tianqi Chen, Gennady Pekhimenko, 2022, MLSys.

D IET C ODE : A UTOMATIC O PTIMIZATION FOR D YNAMIC T ENSOR P ROGRAMS

Cody Hao Yu, Tianqi Chen, Gennady Pekhimenko, 2022 .

Echo: Compiler-based GPU Memory Footprint Reduction for LSTM RNN Training

Gennady Pekhimenko, N. Vijaykumar, Bojian Zheng, 2018, 2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA).

OP-CBIO140912 1553..1560

C. Alkan, Gennady Pekhimenko, O. Mutlu, 2015 .

Hidet: Task Mapping Programming Paradigm for Deep Learning Tensor Programs

Cody Hao Yu, Gennady Pekhimenko, Bojian Zheng, 2022, ArXiv.

A Vector Institute Industry Collaborative Project A Technical Report Harnessing the Power of Natural Language Processing (NLP):

Garth A. Gibson, Jimmy Ba, Gennady Pekhimenko, 2020 .

Tempo: Accelerating Transformer-Based Model Training through Memory Footprint Reduction

Gennady Pekhimenko, Bojian Zheng, Zhanda Zhu, 2022, ArXiv.

Energy-Efficient Data Compression for GPU Memory Systems

Gennady Pekhimenko, O. Mutlu, T. Mowry, 2014 .

GPUPool: A Holistic Approach to Fine-Grained GPU Sharing in the Cloud

Gennady Pekhimenko, N. Vijaykumar, Xiaodan Serina Tan, 2022 .

Gist: Efficient Data Encoding for Deep Neural Network Training

Amar Phanishayee, Lingjia Tang, Jason Mars, 2018, 2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA).

Federated benchmarking of medical artificial intelligence with MedPerf pdf

Satyananda Kashyap, Nicolas Padoy, Grigori Fursin, 2021, Nature Machine Intelligence.

MedPerf: Open Benchmarking Platform for Medical AI using Federated Evaluation (npj Digital Medicine, arxiv)

Jayaraman J. Thiagarajan, Daniel J. Beutel, Micah J. Sheller, 2022 .

Efficient Program Compilation Through Machine Learning Techniques

Gennady Pekhimenko, Angela Demke Brown, Gennady Pekhimenko, 2010, Software Automatic Tuning, From Concepts to State-of-the-Art Results.

TensorDash: Exploiting Sparsity to Accelerate Deep Neural Network Training

Andreas Moshovos, Gennady Pekhimenko, Jorge Albericio, 2020, 2020 53rd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).

Benchmarking and Analyzing Deep Neural Network Training

Amar Phanishayee, Bianca Schroeder, Gennady Pekhimenko, 2018, 2018 IEEE International Symposium on Workload Characterization (IISWC).

ROLLER: Fast and Efficient Tensor Compilation for Deep Learning

Asaf Cidon, Gennady Pekhimenko, Hongyu Zhu, 2022, OSDI.

Priority-based Parameter Propagation for Distributed DNN Training pdf

Gennady Pekhimenko, Garth A. Gibson, Alexandra Fedorova, 2019, SysML.

EcoRNN: Efficient Computing of LSTM RNN Training on GPUs

Bojian Zheng, Akshay Nair, Qiongsi Wu, 2018 .

EcoRNN: Fused LSTM RNN Implementation with Data Layout Optimization pdf

Gennady Pekhimenko, Nandita Vijaykumar, Bojian Zheng, 2018, ArXiv.

BPPSA: Scaling Back-propagation by Parallel Scan Algorithm

Gennady Pekhimenko, Yifan Bai, Shang Wang, 2019, MLSys.

Scaling Back-propagation by Parallel Scan Algorithm pdf

Gennady Pekhimenko, Yifan Bai, Shang Wang, 2019, ArXiv.

Automatic Horizontal Fusion for GPU Kernels pdf

Ao Li, Fan Long, Gennady Pekhimenko, 2020, 2022 IEEE/ACM International Symposium on Code Generation and Optimization (CGO).

Distributed Deep Learning in Open Collaborations pdf

Gennady Pekhimenko, Thomas Wolf, Alexey Bukhtiyarov, 2021, NeurIPS.

Hidet: Task-Mapping Programming Paradigm for Deep Learning Tensor Programs

Cody Hao Yu, Gennady Pekhimenko, Bojian Zheng, 2022, ASPLOS.

IOS: Inter-Operator Scheduler for CNN Acceleration

Song Han, Zhihao Jia, Gennady Pekhimenko, 2020, MLSys.

MLPerf Training Benchmark

Cody A. Coleman, D. Narayanan, Daniel Kang, 2019, MLSys.

Istc-cc Update Istc-cc Research Overview Pillar 1: Specialization

Henggang Cui, Jinliang Wei, Gennady Pekhimenko, 2022 .

Moshpit SGD: Communication-Efficient Decentralized Training on Heterogeneous Unreliable Devices

Eduard A. Gorbunov, Gennady Pekhimenko, Max Ryabinin, 2021, NeurIPS.

Habitat: A Runtime-Based Computational Performance Predictor for Deep Neural Network Training

Gennady Pekhimenko, Geoffrey X. Yu, Pavel Golikov, 2021, USENIX Annual Technical Conference.

FPRaker: A Processing Element For Accelerating Neural Network Training pdf

Andreas Moshovos, Gennady Pekhimenko, Mostafa Mahmoud, 2020, MICRO.

Design-Induced Latency Variation in Modern DRAM Chips: Characterization, Analysis, and Latency Reduction Mechanisms

Rachata Ausavarungnirun, Onur Mutlu, Samira Manabi Khan, 2017, SIGMETRICS.

SoftMC: A Flexible and Practical Open-Source Infrastructure for Enabling Experimental DRAM Studies

Onur Mutlu, Oguz Ergin, Samira Manabi Khan, 2017, 2017 IEEE International Symposium on High Performance Computer Architecture (HPCA).

Understanding Latency Variation in Modern DRAM Chips: Experimental Characterization, Analysis, and Optimization

Onur Mutlu, Tianshi Li, Samira Manabi Khan, 2016, SIGMETRICS.

SoftMC: Practical DRAM Characterization Using an FPGA-Based Infrastructure pdf

Onur Mutlu, Oguz Ergin, Samira Manabi Khan, 2018, ArXiv.

Flexible-Latency DRAM: Understanding and Exploiting Latency Variation in Modern DRAM Chips pdf

Onur Mutlu, Tianshi Li, Gennady Pekhimenko, 2018, ArXiv.

Design-Induced Latency Variation in Modern DRAM Chips

Gennady Pekhimenko, Vivek Seshadri, O. Mutlu, 2016, Proc. ACM Meas. Anal. Comput. Syst..

Reducing DRAM Latency by Exploiting Design-Induced Latency Variation in Modern DRAM Chips pdf

Rachata Ausavarungnirun, Onur Mutlu, Samira Manabi Khan, 2016, ArXiv.

Multi-node Bert-pretraining: Cost-efficient Approach pdf

Xin Li, Jiahuang Lin, Gennady Pekhimenko, 2020, ArXiv.

A case for Core-Assisted Bottleneck Acceleration in GPUs: Enabling flexible data compression with assist warps

Mahmut T. Kandemir, Chita R. Das, Rachata Ausavarungnirun, 2015, 2015 ACM/IEEE 42nd Annual International Symposium on Computer Architecture (ISCA).

Shifted Hamming distance: a fast and accurate SIMD-friendly filter to accelerate alignment verification in read mapping

Onur Mutlu, Gennady Pekhimenko, Carl Kingsford, 2015, Bioinform..

Optimal seed solver: optimizing seed selection in read mapping

Onur Mutlu, Gennady Pekhimenko, Carl Kingsford, 2015, Bioinform..

Janus: Optimizing Memory and Storage Support for Non-Volatile Memory Systems

Gennady Pekhimenko, Samira Khan, Sihang Liu, 2019, 2019 ACM/IEEE 46th Annual International Symposium on Computer Architecture (ISCA).

StreamBox-HBM: Stream Analytics on High Bandwidth Hybrid Memory

Hongyu Miao, Kathryn S. McKinley, Gennady Pekhimenko, 2019, ASPLOS.

PocketTrend: Timely Identification and Delivery of Trending Search Content to Mobile Users

Karin Strauss, Oriana Riva, Gennady Pekhimenko, 2015, WWW.

A Case for Richer Cross-Layer Abstractions: Bridging the Semantic Gap with Expressive Memory

Onur Mutlu, Gennady Pekhimenko, Phillip B. Gibbons, 2018, 2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA).

TiLT: A Time-Centric Approach for Stream Query Optimization and Parallelization

Gennady Pekhimenko, Wei Zhao, Anand Jayarajan, 2023, International Conference on Architectural Support for Programming Languages and Operating Systems.

LifeStream: a high-performance stream processing engine for periodic streams

Gennady Pekhimenko, Anand Jayarajan, K. Hau, 2021, ASPLOS.

TerseCades: Efficient Data Compression in Stream Processing

Peng Huang, Gennady Pekhimenko, Chuanxiong Guo, 2018, USENIX Annual Technical Conference.

The Vision Behind MLPerf: Understanding AI Inference Performance

Carole-Jean Wu, Gennady Pekhimenko, Thomas B. Jablin, 2021, IEEE Micro.

A Runtime-Based Computational Performance Predictor for Deep Neural Network Training

Gennady Pekhimenko, Geoffrey X. Yu, Pavel Golikov, 2021, 2102.00527.

Computational Performance Predictions for Deep Neural Network Training: A Runtime-Based Approach pdf

Gennady Pekhimenko, Geoffrey X. Yu, Pavel Golikov, 2021, ArXiv.

Horizontally Fused Training Array: An Effective Hardware Utilization Squeezer for Training Novel Deep Learning Models

Gennady Pekhimenko, Yuxuan Zheng, Peiming Yang, 2021, MLSys.

Skyline: Interactive In-Editor Computational Performance Profiling for Deep Neural Network Training

Tovi Grossman, Gennady Pekhimenko, Geoffrey X. Yu, 2020, UIST.

Daydream: Accurately Estimating the Efficacy of Optimizations for DNN Training

Amar Phanishayee, Gennady Pekhimenko, Hongyu Zhu, 2020, USENIX Annual Technical Conference.

Speeding up Fourier Neural Operators via Mixed Precision

K. Azizzadenesheli, Anima Anandkumar, Gennady Pekhimenko, 2023, ArXiv.

Adaptive-latency DRAM: Optimizing DRAM timing for the common-case

Gennady Pekhimenko, Vivek Seshadri, O. Mutlu, 2015, 2015 IEEE 21st International Symposium on High Performance Computer Architecture (HPCA).

LifeStream: A High-performance Stream Processing Engine for Waveform Data pdf

Gennady Pekhimenko, Anand Jayarajan, Kimberly Hau, 2020, ArXiv.

RL-Scope: Cross-Stack Profiling for Deep Reinforcement Learning Workloads pdf

Eyal de Lara, Gennady Pekhimenko, Srivatsan Krishnan, 2021, MLSys.

StreamBox: Modern Stream Processing on a Multicore Machine

Hongyu Miao, Kathryn S. McKinley, Gennady Pekhimenko, 2017, USENIX Annual Technical Conference.

EcoRNN: Efficient Computing of LSTM RNN on GPUs

Gennady Pekhimenko, Bojian Zheng, Gennady Pekhimenko, 2018, MICRO 2018.

MLSys: The New Frontier of Machine Learning Systems

Christopher De Sa, Michael I. Jordan, Phillip B. Gibbons, 2019, 1904.03257.

SysML: The New Frontier of Machine Learning Systems

Christopher De Sa, Michael I. Jordan, Phillip B. Gibbons, 2019, ArXiv.

Adaptive-Latency DRAM (AL-DRAM) pdf

Kevin Kai-Wei Chang, Onur Mutlu, Samira Manabi Khan, 2016, ArXiv.

Decoupling GPU Programming Models from Resource Management for Enhanced Programming Ease, Portability, and Performance pdf

Onur Mutlu, Samira Manabi Khan, Gennady Pekhimenko, 2018, ArXiv.

Zorua: Enhancing Programming Ease, Portability, and Performance in GPUs by Decoupling Programming Models from Resource Management pdf

Onur Mutlu, Samira Manabi Khan, Gennady Pekhimenko, 2018, ArXiv.

Flexible-Latency DRAM: Understanding and Exploiting Latency Variation in Modern DRAM Chips

Gennady Pekhimenko, O. Mutlu, Donghyuk Lee, 2018, arXiv.org.

Practical Data Compression for Modern Memory Hierarchies pdf

Gennady Pekhimenko, Gennady Pekhimenko, 2016, ArXiv.

Mitigating the Memory Bottleneck With Approximate Load Value Prediction

Onur Mutlu, Todd C. Mowry, Gennady Pekhimenko, 2016, IEEE Design & Test.

A case for toggle-aware compression for GPU systems

Onur Mutlu, Todd C. Mowry, Gennady Pekhimenko, 2016, 2016 IEEE International Symposium on High Performance Computer Architecture (HPCA).

Habitat: A Runtime-Based Computational Performance Predictor for Deep Neural Network Training

Geoffrey X. Yu, Gennady Pekhimenko, P. Golikov, 2021, USENIX Annual Technical Conference.

Grape: Practical and Efficient Graph-based Executions for Dynamic Deep Neural Networks on GPUs

Gennady Pekhimenko, Bojian Zheng, 2022 .

A Framework for Accelerating Bottlenecks in GPU Execution with Assist Warps pdf

Mahmut T. Kandemir, Chita R. Das, Rachata Ausavarungnirun, 2016, ArXiv.

Speeding up Fourier Neural Operators via Mixed Precision

K. Azizzadenesheli, Anima Anandkumar, Gennady Pekhimenko, 2023, ArXiv.

The Synergy of Speculative Decoding and Batching in Serving Large Language Models

Gennady Pekhimenko, Qidong Su, Christina Giannoula, 2023, ArXiv.

BPPSA: Scaling Back-propagation by Parallel Scan Algorithm

Gennady Pekhimenko, Yifan Bai, Shang Wang, 2019, MLSys.

NVOverlay: Enabling Efficient and Scalable High-Frequency Snapshotting to NVM

Todd C. Mowry, Gennady Pekhimenko, Michael A. Kozuch, 2021, 2021 ACM/IEEE 48th Annual International Symposium on Computer Architecture (ISCA).

Grape: Practical and Efficient Graphed Execution for Dynamic Deep Neural Networks on GPUs

Gennady Pekhimenko, Bojian Zheng, Jie Wang, 2023, MICRO.

Tempo: Accelerating Transformer-Based Model Training through Memory Footprint Reduction

Gennady Pekhimenko, Bojian Zheng, Zhanda Zhu, 2022, NeurIPS.

Boveda: Building an On-Chip Deep Learning Memory Hierarchy Brick by Brick

Alberto Delmas Lascorz, Gennady Pekhimenko, Sayeh Sharify, 2021, MLSys.

Design-Induced Latency Variation in Modern DRAM Chips

Gennady Pekhimenko, Vivek Seshadri, O. Mutlu, 2017, Proc. ACM Meas. Anal. Comput. Syst..

Simultaneous Multi-Layer Access

Gennady Pekhimenko, O. Mutlu, Donghyuk Lee, 2016, ACM Trans. Archit. Code Optim..