Lost In Translation: Exposing Hidden Compiler Optimization Opportunities

Abstract Existing iterative compilation and machine learning-based optimization techniques have been proven very successful in achieving better optimizations than the standard optimization levels of a compiler. However, they were not engineered to support the tuning of a compiler’s optimizer as part of the compiler’s daily development cycle. In this paper, we first establish the required properties that a technique must exhibit to enable such tuning. We then introduce an enhancement to the classic nightly routine testing of compilers, which exhibits all the required properties and thus is capable of driving the improvement and tuning of the compiler’s common optimizer. This is achieved by leveraging resource usage and compilation information collected while systematically exploiting prefixes of the transformations applied at standard optimization levels. Experimental evaluation using the LLVM v6.0.1 compiler demonstrated that the new approach was able to reveal hidden cross-architecture and architecture-dependent potential optimizations on two popular processors: the Intel i5-6300U and the Arm Cortex-A53-based Broadcom BCM2837 used in the Raspberry Pi 3B+. As a case study, we demonstrate how the insights from our approach enabled us to identify and remove a significant shortcoming of the CFG simplification pass of the LLVM v6.0.1 compiler.

Download Full-text

The multilevel ant stigmergy algorithm for numerical optimization

Facta universitatis - series Electronics and Energetics ◽

10.2298/fuee0602247k ◽

2006 ◽

Vol 19 (2) ◽

pp. 247-260 ◽

Cited By ~ 2

Author(s):

Peter Korosec ◽

Jurij Silc

Keyword(s):

Differential Evolution ◽

Optimization Algorithm ◽

Numerical Optimization ◽

Optimization Techniques ◽

Industrial Case Study ◽

New Approach ◽

Collective Work ◽

Better Than

The Multilevel Ant Stigmergy Algorithm (MASA) is a new approach to solving multi-parameter problems based on stigmergy, a type of collective work that can be observed in nature. In this paper we evaluate the performance of MASA regarding its applicability as numerical optimization techniques. The evaluation is performed with several widely used benchmarks functions, as well as on an industrial case study. We also compare the MASA with Differential Evolution, well-known numerical optimization algorithm. The average solution obtained with the MASA was better than a solution recently found using Differential Evolution.

Download Full-text

Revitalizing Ways of Korean Traditional Markets through A Overseas Case Study: A New Approach to Revitalize Korean Traditional Markets

Korea Association of Business Education ◽

10.23839/kabe.2018.33.3.153 ◽

2018 ◽

Vol 33 (3) ◽

pp. 153-173

Author(s):

Heung－Seob Chang

Keyword(s):

New Approach ◽

Traditional Markets

Download Full-text

Lost in Translation : Revelations 2:12.17:A Case Study for Idioms in Translation

Journal of Biblical Text Research ◽

10.28977/jbtr.2009.4.24.249 ◽

2009 ◽

Vol 24 ◽

pp. 249-282

Author(s):

Thomas Kaut

Keyword(s):

Lost In Translation

Download Full-text

Benefit Corporations Approach to Environmental, Social and Governance Disclosure: A Focus on Italy

Entrepreneurship Research Journal ◽

10.1515/erj-2019-0318 ◽

2020 ◽

Vol 10 (4) ◽

Author(s):

Mauro Sciarelli ◽

Silvia Cosimato ◽

Giovanni Landi

Keyword(s):

Qualitative Method ◽

Positive Impact ◽

Corporate Structure ◽

New Approach ◽

For Profit ◽

Benefit Corporation ◽

Benefit Corporations ◽

Multiple Case ◽

Governance Disclosure

AbstractOver the last decades, Benefit Corporations arouse as a new corporate structure, alternative to traditional ones and pointing to offer a new approach to the management of business and sustainability issues. These companies' activities are statutory aimed at bridging for-profit and no-profit activities; thus, they intentionally and statutory pursue economic purposes together with social and environmental ones, to create a positive impact on economy, society and environment. Even though, Italian and other national laws set some specific disclosure duties for Benefit Corporations, especially in terms of Environmental, Social and Governance (ESG) issues, the literature still calls for further research on the topic. Therefore, this paper is aimed at contributing to bridge this gap, investigating the way Italian Benefit Corporations approach ESG disclosure. To this end, an exploratory analysis has been conducted, implementing a qualitative method, based on a multiple case study strategy. Even though the descriptive nature of the study, the achieved findings pointed out that the Benefit Corporation structure not necessarily implies a better approach to ESG.

Download Full-text

Incorporating lines of evidence from New Approach Methodologies (NAMs) to reduce uncertainties in a category based read-across: A case study for repeated dose toxicity

Regulatory Toxicology and Pharmacology ◽

10.1016/j.yrtph.2020.104855 ◽

2021 ◽

Vol 120 ◽

pp. 104855

Author(s):

Cynthia B. Pestana ◽

James W. Firman ◽

Mark T.D. Cronin

Keyword(s):

Repeated Dose ◽

New Approach ◽

Repeated Dose Toxicity ◽

Lines Of Evidence

Download Full-text

Lightweight Blockchain Processing. Case Study: Scanned Document Tracking on Tezos Blockchain

Applied Sciences ◽

10.3390/app11157169 ◽

2021 ◽

Vol 11 (15) ◽

pp. 7169

Author(s):

Mohamed Allouche ◽

Tarek Frikha ◽

Mihai Mitrea ◽

Gérard Memmi ◽

Faten Chaabane

Keyword(s):

Load Balancing ◽

Relative Error ◽

Execution Time ◽

General Purpose ◽

Experimental Results ◽

Raspberry Pi ◽

Embedded Platform ◽

Memory Resources ◽

Processing Solution

To bridge the current gap between the Blockchain expectancies and their intensive computation constraints, the present paper advances a lightweight processing solution, based on a load-balancing architecture, compatible with the lightweight/embedding processing paradigms. In this way, the execution of complex operations is securely delegated to an off-chain general-purpose computing machine while the intimate Blockchain operations are kept on-chain. The illustrations correspond to an on-chain Tezos configuration and to a multiprocessor ARM embedded platform (integrated into a Raspberry Pi). The performances are assessed in terms of security, execution time, and CPU consumption when achieving a visual document fingerprint task. It is thus demonstrated that the advanced solution makes it possible for a computing intensive application to be deployed under severely constrained computation and memory resources, as set by a Raspberry Pi 3. The experimental results show that up to nine Tezos nodes can be deployed on a single Raspberry Pi 3 and that the limitation is not derived from the memory but from the computation resources. The execution time with a limited number of fingerprints is 40% higher than using a classical PC solution (value computed with 95% relative error lower than 5%).

Download Full-text

Investigating the Potential of Network Optimization for a Constrained Object Detection Problem

Journal of Imaging ◽

10.3390/jimaging7040064 ◽

2021 ◽

Vol 7 (4) ◽

pp. 64

Author(s):

Tanguy Ophoff ◽

Cédric Gullentops ◽

Kristof Van Beeck ◽

Toon Goedemé

Keyword(s):

Computational Complexity ◽

Object Detection ◽

Network Optimization ◽

Real Life ◽

Optimization Techniques ◽

Training Data ◽

Single Shot ◽

Standard Object ◽

Number Of Classes

Object detection models are usually trained and evaluated on highly complicated, challenging academic datasets, which results in deep networks requiring lots of computations. However, a lot of operational use-cases consist of more constrained situations: they have a limited number of classes to be detected, less intra-class variance, less lighting and background variance, constrained or even fixed camera viewpoints, etc. In these cases, we hypothesize that smaller networks could be used without deteriorating the accuracy. However, there are multiple reasons why this does not happen in practice. Firstly, overparameterized networks tend to learn better, and secondly, transfer learning is usually used to reduce the necessary amount of training data. In this paper, we investigate how much we can reduce the computational complexity of a standard object detection network in such constrained object detection problems. As a case study, we focus on a well-known single-shot object detector, YoloV2, and combine three different techniques to reduce the computational complexity of the model without reducing its accuracy on our target dataset. To investigate the influence of the problem complexity, we compare two datasets: a prototypical academic (Pascal VOC) and a real-life operational (LWIR person detection) dataset. The three optimization steps we exploited are: swapping all the convolutions for depth-wise separable convolutions, perform pruning and use weight quantization. The results of our case study indeed substantiate our hypothesis that the more constrained a problem is, the more the network can be optimized. On the constrained operational dataset, combining these optimization techniques allowed us to reduce the computational complexity with a factor of 349, as compared to only a factor 9.8 on the academic dataset. When running a benchmark on an Nvidia Jetson AGX Xavier, our fastest model runs more than 15 times faster than the original YoloV2 model, whilst increasing the accuracy by 5% Average Precision (AP).

Download Full-text

Towards a New Approach for Developing Patient Flow in Emergency Department: Sahloul and Hached Hospitals Case Study

2020 IEEE 13th International Colloquium of Logistics and Supply Chain Management (LOGISTIQUA) ◽

10.1109/logistiqua49782.2020.9353899 ◽

2020 ◽

Author(s):

Imen Hamzaoui ◽

Aida Bouzir ◽

Saloua Benammou

Keyword(s):

Emergency Department ◽

Patient Flow ◽

New Approach

Download Full-text

GLOSSING THE GLOSS: READING PETER LOMBARD’SCOLLECTANEAON THE PAULINE EPISTLES AS A HISTORICAL ACT

Traditio ◽

10.1017/tdo.2018.8 ◽

2018 ◽

Vol 73 ◽

pp. 83-116

Author(s):

PETER O'HAGAN

Keyword(s):

Early Modern ◽

Biblical Text ◽

Classroom Activity ◽

New Approach ◽

Central Focus ◽

Recent Scholarship ◽

Pauline Epistles ◽

History Of ◽

Extended Analysis

Peter Lombard's influential commentary on the Pauline Epistles, theCollectanea in omnes divi Pauli epistolas,has received little extended analysis in scholarly literature, despite its recognized importance both in its own right and as key for the development of hisSentences.This article presents a new approach to studying theCollectaneaby analyzing how Lombard's commentary builds on theGlossa “Ordinaria”on the Pauline Epistles. The article argues for treating theCollectaneaas a “historical act,” focusing on how Lombard engages with the biblical text and with authoritative sources within which he encounters the same biblical text embedded. The article further argues for the necessity of turning to the manuscripts of both theCollectaneaand theGlossa,rather than continuing to rely on inadequate early modern printed editions or thePatrologia Latina.The article then uses Lombard's discussion of faith at Romans 1:17 as a case study, demonstrating the way in which Lombard begins from theGlossa,clarifies its ambiguities, and moves his analysis forward through his use of otherauctoritatesand theologicalquaestiones.A comparison with Lombard's treatment of faith in theSentenceshighlights the close links between Lombard's biblical lectures and this later work. The article concludes by arguing that scholastic biblical exegesis and theology should be treated as primarily a classroom activity, with the glossed Bible as the central focus. Discussion of Lombard's work should draw on much recent scholarship that has begun to uncover the layers of orality within the textual history of scholastic works.

Download Full-text