{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/110468"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/110468","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"ApproxHPVM: A retargetable compiler framework for accuracy-aware optimizations","abstract":"With the increasing need for machine learning and data processing near the edge, software stacks and compilers must provide optimizations for alleviating the computational burden on low-end edge devices. Approximate computing can help bridge the gap between increasing computational demands and limited compute power on such devices. We present ApproxHPVM, a portable optimizing compiler and runtime system that enables flexible, optimized use of multiple software and hardware approximations in a unified easy-to-use framework. ApproxHPVM uses a portable compiler IR and compiler analyses that are designed to enable accuracy-aware performance and energy tuning on heterogeneous systems with multiple compute units and approximation methods. ApproxHPVM automatically translates end-to-end application-level quality metrics into accuracy requirements for individual operations. ApproxHPVM uses a hardware-agnostic accuracy-tuning phase to do this translation that provides greater portability across heterogeneous hardware platforms. ApproxHPVM incorporates three main components: (a) a compiler IR with hardware-agnostic approximation metrics, (b) a hardware-agnostic accuracy-tuning phase to identify error-tolerant computations, and (c) an accuracy-aware hardware scheduler that maps error-tolerant computations to approximate hardware components. As ApproxHPVM does not incorporate any hardware-specific knowledge as part of the IR, it can serve as a portable virtual ISA that can be shipped to all kinds of hardware platforms. We evaluate ApproxHPVM on 9 benchmarks from the deep learning domain and 5 image-processing benchmarks. Our results show that our framework can offload chunks of approximable computations to special-purpose accelerators that provide significant gains in performance and energy, while staying within user-specified application-level quality metrics with high probability. Across the 14 benchmarks, we observe from 1-9x performance speedups and 1.1-11.3x energy reduction for very small reductions in accuracy. ApproxTuner extends ApproxHPVM with a flexible system for dynamic approximation tuning. The key contribution in ApproxTuner is a novel three-phase approach to approximation-tuning that consists of development-time, install-time, and run-time phases. Our approach decouples tuning hardware-independent and hardware-specific approximations, thus providing retargetability across devices. To enable efficient autotuning of approximation choices, we present a novel accuracy-aware tuning technique called predictive approximation-tuning. It can optimize the application during development-time and can also refine the optimization with (previously unknown) hardware-specific approximations at install time. We evaluate ApproxTuner across 11 benchmarks from deep learning and image processing domains. For the evaluated convolutional neural networks, we show that using only hardware-independent approximation choices provides a mean speedup of 2.2x (max 2.7x) on GPU, and 1.4x mean speedup (max 1.9x) on the CPU, while staying within 2 percentage points of inference accuracy loss. For two different accuracy-prediction models, our predictive tuning strategy speeds up tuning by 13.7x and 17.9x compared to conventional empirical tuning while achieving comparable benefits.","abstract_html":"With the increasing need for machine learning and data processing near the edge, software stacks and compilers must provide optimizations for alleviating the computational burden on low-end edge devices. Approximate computing can help bridge the gap between increasing computational demands and limited compute power on such devices. We present ApproxHPVM, a portable optimizing compiler and runtime system that enables flexible, optimized use of multiple software and hardware approximations in a unified easy-to-use framework. ApproxHPVM uses a portable compiler IR and compiler analyses that are designed to enable accuracy-aware performance and energy tuning on heterogeneous systems with multiple compute units and approximation methods. ApproxHPVM automatically translates end-to-end application-level quality metrics into accuracy requirements for individual operations. ApproxHPVM uses a hardware-agnostic accuracy-tuning phase to do this translation that provides greater portability across heterogeneous hardware platforms. ApproxHPVM incorporates three main components: (a) a compiler IR with hardware-agnostic approximation metrics, (b) a hardware-agnostic accuracy-tuning phase to identify error-tolerant computations, and (c) an accuracy-aware hardware scheduler that maps error-tolerant computations to approximate hardware components. As ApproxHPVM does not incorporate any hardware-specific knowledge as part of the IR, it can serve as a portable virtual ISA that can be shipped to all kinds of hardware platforms. We evaluate ApproxHPVM on 9 benchmarks from the deep learning domain and 5 image-processing benchmarks. Our results show that our framework can offload chunks of approximable computations to special-purpose accelerators that provide significant gains in performance and energy, while staying within user-specified application-level quality metrics with high probability. Across the 14 benchmarks, we observe from 1-9x performance speedups and 1.1-11.3x energy reduction for very small reductions in accuracy. ApproxTuner extends ApproxHPVM with a flexible system for dynamic approximation tuning. The key contribution in ApproxTuner is a novel three-phase approach to approximation-tuning that consists of development-time, install-time, and run-time phases. Our approach decouples tuning hardware-independent and hardware-specific approximations, thus providing retargetability across devices. To enable efficient autotuning of approximation choices, we present a novel accuracy-aware tuning technique called predictive approximation-tuning. It can optimize the application during development-time and can also refine the optimization with (previously unknown) hardware-specific approximations at install time. We evaluate ApproxTuner across 11 benchmarks from deep learning and image processing domains. For the evaluated convolutional neural networks, we show that using only hardware-independent approximation choices provides a mean speedup of 2.2x (max 2.7x) on GPU, and 1.4x mean speedup (max 1.9x) on the CPU, while staying within 2 percentage points of inference accuracy loss. For two different accuracy-prediction models, our predictive tuning strategy speeds up tuning by 13.7x and 17.9x compared to conventional empirical tuning while achieving comparable benefits.","abstract_has_math":false,"creators":["Sharif, Hashim"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Adve, Vikram","Adve, Sarita","Misailovic, Sasa","Amarasinghe, Saman","Hoffmann, Henry"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2021,"date_issued":"2021-09-17T01:10:49Z","date_published":"2021-09-17T01:10:49Z","updated_at":"2026-07-22T22:24:50Z","subjects":["Compilers","Heterogeneous Systems","Approximate Computing","Approximation Tuning","Deep Learning","Robotics"],"languages":["en"],"rights":["Copyright 2021 Hashim Sharif"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/110468","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Adve, Vikram","Adve, Sarita","Misailovic, Sasa","Amarasinghe, Saman","Hoffmann, Henry"]},{"key":"dc:creator","label":"Author","values":["Sharif, Hashim"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2021-09-17T01:10:49Z","2021-04-14","2021-05"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Compilers","Heterogeneous Systems","Approximate Computing","Approximation Tuning","Deep Learning","Robotics"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2021 Hashim Sharif"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/110468"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["With the increasing need for machine learning and data processing near the edge, software stacks and compilers must provide optimizations for alleviating the computational burden on low-end edge devices. Approximate computing can help bridge the gap between increasing computational demands and limited compute power on such devices. We present ApproxHPVM, a portable optimizing compiler and runtime system that enables flexible, optimized use of multiple software and hardware approximations in a unified easy-to-use framework. ApproxHPVM uses a portable compiler IR and compiler analyses that are designed to enable accuracy-aware performance and energy tuning on heterogeneous systems with multiple compute units and approximation methods. ApproxHPVM automatically translates end-to-end application-level quality metrics into accuracy requirements for individual operations. ApproxHPVM uses a hardware-agnostic accuracy-tuning phase to do this translation that provides greater portability across heterogeneous hardware platforms. ApproxHPVM incorporates three main components: (a) a compiler IR with hardware-agnostic approximation metrics, (b) a hardware-agnostic accuracy-tuning phase to identify error-tolerant computations, and (c) an accuracy-aware hardware scheduler that maps error-tolerant computations to approximate hardware components. As ApproxHPVM does not incorporate any hardware-specific knowledge as part of the IR, it can serve as a portable virtual ISA that can be shipped to all kinds of hardware platforms. We evaluate ApproxHPVM on 9 benchmarks from the deep learning domain and 5 image-processing benchmarks. Our results show that our framework can offload chunks of approximable computations to special-purpose accelerators that provide significant gains in performance and energy, while staying within user-specified application-level quality metrics with high probability. Across the 14 benchmarks, we observe from 1-9x performance speedups and 1.1-11.3x energy reduction for very small reductions in accuracy. ApproxTuner extends ApproxHPVM with a flexible system for dynamic approximation tuning. The key contribution in ApproxTuner is a novel three-phase approach to approximation-tuning that consists of development-time, install-time, and run-time phases. Our approach decouples tuning hardware-independent and hardware-specific approximations, thus providing retargetability across devices. To enable efficient autotuning of approximation choices, we present a novel accuracy-aware tuning technique called predictive approximation-tuning. It can optimize the application during development-time and can also refine the optimization with (previously unknown) hardware-specific approximations at install time. We evaluate ApproxTuner across 11 benchmarks from deep learning and image processing domains. For the evaluated convolutional neural networks, we show that using only hardware-independent approximation choices provides a mean speedup of 2.2x (max 2.7x) on GPU, and 1.4x mean speedup (max 1.9x) on the CPU, while staying within 2 percentage points of inference accuracy loss. For two different accuracy-prediction models, our predictive tuning strategy speeds up tuning by 13.7x and 17.9x compared to conventional empirical tuning while achieving comparable benefits.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2021-09-16 without embargo terms","The student, Hashim Sharif, accepted the attached license on 2021-04-13 at 17:34.","The student, Hashim Sharif, submitted this Dissertation for approval on 2021-04-13 at 17:56.","This Dissertation was approved for publication on 2021-04-14 at 15:36.","DSpace SAF Submission Ingestion Package generated from Vireo submission #16314 on 2021-09-16 at 16:41:13","Made available in DSpace on 2021-09-17T01:10:49Z (GMT). No. of bitstreams: 3 SHARIF-DISSERTATION-2021.pdf: 4515981 bytes, checksum: a5af485058e1762dba3ffcb8cd465c13 (MD5) LICENSE.txt: 4210 bytes, checksum: d2795eb112003d1775257fb13b004fa0 (MD5) PROQUEST_LICENSE.txt: 4556 bytes, checksum: ff22f6f87a6ced6539a7d8e059abcf46 (MD5) Previous issue date: 2021-04-14"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["ApproxHPVM: A retargetable compiler framework for accuracy-aware optimizations"]}]}],"canonical_facts":{"dc:contributor":["Adve, Vikram","Adve, Sarita","Misailovic, Sasa","Amarasinghe, Saman","Hoffmann, Henry"],"dc:creator":["Sharif, Hashim"],"dc:date":["2021-09-17T01:10:49Z","2021-04-14","2021-05"],"dc:description":["With the increasing need for machine learning and data processing near the edge, software stacks and compilers must provide optimizations for alleviating the computational burden on low-end edge devices. Approximate computing can help bridge the gap between increasing computational demands and limited compute power on such devices. We present ApproxHPVM, a portable optimizing compiler and runtime system that enables flexible, optimized use of multiple software and hardware approximations in a unified easy-to-use framework. ApproxHPVM uses a portable compiler IR and compiler analyses that are designed to enable accuracy-aware performance and energy tuning on heterogeneous systems with multiple compute units and approximation methods. ApproxHPVM automatically translates end-to-end application-level quality metrics into accuracy requirements for individual operations. ApproxHPVM uses a hardware-agnostic accuracy-tuning phase to do this translation that provides greater portability across heterogeneous hardware platforms. ApproxHPVM incorporates three main components: (a) a compiler IR with hardware-agnostic approximation metrics, (b) a hardware-agnostic accuracy-tuning phase to identify error-tolerant computations, and (c) an accuracy-aware hardware scheduler that maps error-tolerant computations to approximate hardware components. As ApproxHPVM does not incorporate any hardware-specific knowledge as part of the IR, it can serve as a portable virtual ISA that can be shipped to all kinds of hardware platforms. We evaluate ApproxHPVM on 9 benchmarks from the deep learning domain and 5 image-processing benchmarks. Our results show that our framework can offload chunks of approximable computations to special-purpose accelerators that provide significant gains in performance and energy, while staying within user-specified application-level quality metrics with high probability. Across the 14 benchmarks, we observe from 1-9x performance speedups and 1.1-11.3x energy reduction for very small reductions in accuracy. ApproxTuner extends ApproxHPVM with a flexible system for dynamic approximation tuning. The key contribution in ApproxTuner is a novel three-phase approach to approximation-tuning that consists of development-time, install-time, and run-time phases. Our approach decouples tuning hardware-independent and hardware-specific approximations, thus providing retargetability across devices. To enable efficient autotuning of approximation choices, we present a novel accuracy-aware tuning technique called predictive approximation-tuning. It can optimize the application during development-time and can also refine the optimization with (previously unknown) hardware-specific approximations at install time. We evaluate ApproxTuner across 11 benchmarks from deep learning and image processing domains. For the evaluated convolutional neural networks, we show that using only hardware-independent approximation choices provides a mean speedup of 2.2x (max 2.7x) on GPU, and 1.4x mean speedup (max 1.9x) on the CPU, while staying within 2 percentage points of inference accuracy loss. For two different accuracy-prediction models, our predictive tuning strategy speeds up tuning by 13.7x and 17.9x compared to conventional empirical tuning while achieving comparable benefits.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2021-09-16 without embargo terms","The student, Hashim Sharif, accepted the attached license on 2021-04-13 at 17:34.","The student, Hashim Sharif, submitted this Dissertation for approval on 2021-04-13 at 17:56.","This Dissertation was approved for publication on 2021-04-14 at 15:36.","DSpace SAF Submission Ingestion Package generated from Vireo submission #16314 on 2021-09-16 at 16:41:13","Made available in DSpace on 2021-09-17T01:10:49Z (GMT). No. of bitstreams: 3 SHARIF-DISSERTATION-2021.pdf: 4515981 bytes, checksum: a5af485058e1762dba3ffcb8cd465c13 (MD5) LICENSE.txt: 4210 bytes, checksum: d2795eb112003d1775257fb13b004fa0 (MD5) PROQUEST_LICENSE.txt: 4556 bytes, checksum: ff22f6f87a6ced6539a7d8e059abcf46 (MD5) Previous issue date: 2021-04-14"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/110468"],"dc:language":["en"],"dc:rights":["Copyright 2021 Hashim Sharif"],"dc:subject":["Compilers","Heterogeneous Systems","Approximate Computing","Approximation Tuning","Deep Learning","Robotics"],"dc:title":["ApproxHPVM: A retargetable compiler framework for accuracy-aware optimizations"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:24:50Z"}