AI-Driven Optimization of Compiler Pipelines for Heterogeneous Processing Architectures
DOI:
https://doi.org/10.63282/3117-5481/AIJCST-V6I5P102Keywords:
AI-Driven Compilation, Heterogeneous Computing, Reinforcement Learning, LLVM, Pipeline Optimization, Performance PortabilityAbstract
Heterogeneous computing systems are growing in prominence in modern applications with heterogeneous processing architectures (HPAs)- CPUs, GPUs, FPGAs, TPUs and domain-specific accelerators being added to fulfill the increasing performance and energy requirements. The compilation process of HPAs is however, still a complicated bottleneck with architectural divergence, mismatchment of multiple memory hierarchies and different models of execution. The conventional static compilers have a hard time in globally optimizing and are specifically aggressive common sense heuristics that do not generalize to dynamic loads. The introduction of Artificial Intelligence (AI) presents new possibilities of predictive compiler pipeline, making it possible to make intelligent optimization choices, including instruction timing, register assignment, kernel division, and mapping devices. The paper is aimed at offering an A compiler-based Optimization Framework (AIOF) that implements machine learning (ML) and deep reinforcement learning (DRL) into the compiler transformation cycle. AIOF automatically discovers the best optimization passes, and decreases the compilation time and enhances performance portability. We introduce a unified pipeline with (1) dynamics feature selection on intermediate representation (IRs), (2) decision models to transform sequences learned, and (3) feedback-based performance assessor based on the use of both static and runtime profiling. Experimental tests show average execution time improvements of 28-55% with heterogeneous computing platforms, and 35% in compilation latency reduces when compared to standard levels of performance of the LLVM based optimization (-O2/-O3). According to our findings, there was an improvement in the level of computational efficiency, workload scalability, and energy sustainability.The contribution that this work makes to the future of autonomous compilers is a scalable evolution framework that will be used in further autonomous compilers, thereby pushing the state of the art forward to self-optimization software ecosystems capable of fully leveraging the hardware diversity
References
[1] P. B. Schneck, “A Survey of Compiler Optimization Techniques,” NASA/ITR, 1972. NASA Technical Reports Server+1
[2] F. M. Q. Pereira, “A Survey on Register Allocation,” UCLA, 2008. compilers.cs.ucla.edu
[3] Z. Wang and M. O’Boyle, “Machine Learning in Compiler Optimisation,” Proc. IEEE, 2018. Research at Edinburgh+1
[4] A. H. Ashouri, W. Killian, J. Cavazos, G. Palermo & C. Silvano, “A Survey on Compiler Autotuning using Machine Learning,” SpringerBriefs in Applied Sciences and Technology, 2018. ResearchGate+1
[5] C. Cummins, “Machine Learning in Compilers: Retrospective and Future,” 2020. chriscummins.cc
[6] H. Shahzad, “Reinforcement Learning Strategies for Compiler Optimisation,” 2022. Boston University
[7] Google Research Blog, “MLGO: A Machine Learning Guided Compiler Optimizations Framework,” 2022. research.google
[8] M. Li et al., “The Deep Learning Compiler: A Comprehensive Survey,” arXiv, 2020. arXiv
[9] R. Maruthamuthu, “Advancements in Compiler Design and Optimization,” E3S Web of Conferences, 2023. E3S Conferences
[10] U. K. R. Bondhugula, “Effective Automatic Parallelization and Locality Optimisation via the Polyhedral Model,” IIT IISc Thesis, 2008. CSA - IISc Bangalore
[11] R. Castañeda Lozano & C. Schulte, “Survey on Combinatorial Register Allocation and Instruction Scheduling,” arXiv, 2014. arXiv
[12] Mohanarajesh Kommineni. Revanth Parvathi. (2013) Risk Analysis for Exploring the Opportunities in Cloud Outsourcing.
[13] Enabling Mission-Critical Communication via VoLTE for Public Safety Networks - Varinder Kumar Sharma - IJAIDR Volume 10, Issue 1, January-June 2019. DOI 10.71097/IJAIDR.v10.i1.1539
[14] Aragani, Venu Madhav and Maroju, Praveen Kumar and Mudunuri, Lakshmi Narasimha Raju, Efficient Distributed Training through Gradient Compression with Sparsification and Quantization Techniques (September 29, 2021). Available at SSRN: https://ssrn.com/abstract=5022841 or http://dx.doi.org/10.2139/ssrn.5022841
[15] P. K. Maroju, "Empowering Data-Driven Decision Making: The Role of Self-Service Analytics and Data Analysts in Modern Organization Strategies," International Journal of Innovations in Applied Science and Engineering (IJIASE), vol. 7, Aug. 2021.
[16] Lakshmi Narasimha Raju Mudunuri, “AI Powered Supplier Selection: Finding the Perfect Fit in Supply Chain Management”, IJIASE, January-December 2021, Vol 7; 211-231.
[17] Kommineni, M. "Explore Knowledge Representation, Reasoning, and Planning Techniques for Building Robust and Efficient Intelligent Systems." International Journal of Inventions in Engineering & Science Technology 7.2 (2021): 105- 114.
[18] Thallam, N. S. T. (2021). Privacy-Preserving Data Analytics in the Cloud: Leveraging Homomorphic Encryption for Big Data Security. Journal of Scientific and Engineering Research, 8(12), 331-337.
[19] Kanji, R. K. (2021). Federated data governance framework for ensuring quality-assured data sharing and integration in hybrid cloud-based data warehouse ecosystems through advanced ETL/ELT techniques. International Journal of Computer Techniques, 8(3), 1-9.
[20] Reinforcement Learning Applications in Self Organizing Networks - Varinder Kumar Sharma - IJIRCT Volume 7 Issue 1, January-2021. DOI: https://doi.org/10.5281/zenodo.17062920
[21] Thirunagalingam, A. (2022). Enhancing Data Governance Through Explainable AI: Bridging Transparency and Automation. Available at SSRN 5047713.
[22] P. K. Maroju, "Conversational AI for Personalized Financial Advice in the BFSI Sector," International Journal of Innovations in Applied Sciences and Engineering, vol. 8, no.2, pp. 156–177, Nov. 2022.
[23] Kulasekhara Reddy Kotte. 2022. ACCOUNTS PAYABLE AND SUPPLIER RELATIONSHIPS: OPTIMIZING PAYMENT CYCLES TO ENHANCE VENDOR PARTNERSHIPS. International Journal of Advances in Engineering Research , 24(6), PP – 14-24, https://www.ijaer.com/admin/upload/02%20Kulasekhara%20Reddy%20Kotte%2001468.pdf
[24] Gopi Chand Vegineni. 2022. Intelligent UI Designs for State Government Applications: Fostering Inclusion without AI and ML, Journal of Advances in Developmental Research, 13(1), PP – 1-13, https://www.ijaidr.com/research-paper.php?id=1454
[25] Hullurappa, M. (2022). The Role of Explainable AI in Building Public Trust: A Study of AI-Driven Public Policy Decisions. International Transactions in Artificial Intelligence, 6.
[26] Bhagath Chandra Chowdari Marella, “Driving Business Success: Harnessing Data Normalization and Aggregation for Strategic Decision-Making”, International Journal of INTELLIGENT SYSTEMS AND APPLICATIONS IN ENGINEERING, vol. 10, no.2, pp. 308 – 317, 2022. https://ijisae.org/index.php/IJISAE/issue/view/87
[27] Mohanarajesh Kommineni. (2022/11/28). Investigating High-Performance Computing Techniques For Optimizing And Accelerating Ai Algorithms Using Quantum Computing And Specialized Hardware. International Journal Of Innovations In Scientific Engineering. 16. 66-80. (Ijise) 2022.
[28] Naga Surya Teja Thallam. (2022). Enhancing Security in Distributed Systems Using Bastion Hosts, NAT Gateways, and Network ACLs. International Scientific Journal of Engineering and Management, 1(1).
[29] Thallam, N. S. T. (2022). Sustainable Cloud Computing: Reducing Carbon Footprint in Large-Scale Cloud Infrastructures. Journal of Scientific and Engineering Research, 9(1), 217-224.
[30] Garg, A. (2022). Unified Framework of Blockchain and AI for Business Intelligence in Modern Banking . International Journal of Emerging Research in Engineering and Technology, 3(4), 32-42. https://doi.org/10.63282/3050-922X.IJERET-V3I4P105
[31] Kanji, R. K. (2022). A Unified Data Warehouse Architecture for Multi-Source Forest Inventory Integration and Automated Remote Sensing Analysis. Sarcouncil Journal of Engineering and Computer Sciences, 1, 10-16.
[32] Varinder Kumar Sharma - AI-Based Anomaly Detection for 5G Core and RAN Components - International Journal of Scientific Research in Engineering and Management (IJSREM) Volume: 06 Issue: 01 | Jan-2022 .DOI: 10.55041/IJSREM11453
[33] Thirunagalingam, A. (2023). Improving Automated Data Annotation with Self-Supervised Learning: A Pathway to Robust AI Models Vol. 7, No. 7,(2023) ITAI. International Transactions in Artificial Intelligence, 7(7).
[34] Praveen Kumar Maroju, "Optimizing Mortgage Loan Processing in Capital Markets: A Machine Learning Approach, " International Journal of Innovations in Scientific Engineering, 17(1), PP. 36-55 , April 2023.
[35] P. K. Maroju, "Leveraging Machine Learning for Customer Segmentation and Targeted Marketing in BFSI," International Transactions in Artificial Intelligence, vol. 7, no. 7, pp. 1-20, Nov. 2023
[36] Kulasekhara Reddy Kotte. 2023. Leveraging Digital Innovation for Strategic Treasury Management: Blockchain, and Real-Time Analytics for Optimizing Cash Flow and Liquidity in Global Corporation. International Journal of Interdisciplinary Finance Insights, 2(2), PP - 1 - 17, https://injmr.com/index.php/ijifi/article/view/186/45
[37] Mudunuri L.N.R.; (December, 2023); “AI-Driven Inventory Management: Never Run Out, Never Overstock”; International Journal of Advances in Engineering Research; Vol 26, Issue 6; 24-36 Sudheer Panyaram, (2023), AI-Powered Framework for Operational Risk Management in the Digital Transformation of Smart Enterprises.
[38] Hullurappa, M. (2023). Intelligent Data Masking: Using GANs to Generate Synthetic Data for Privacy-Preserving Analytics. International Journal of Inventions in Engineering & Science Technology, 9, 9.
[39] B. C. C. Marella, “Data Synergy: Architecting Solutions for Growth and Innovation,” International Journal of Innovative Research in Computer and Communication Engineering, vol. 11, no. 9, pp. 10551–10560, Sep. 2023.
[40] Bhagath Chandra Chowdari Marella, “Scalable Generative AI Solutions for Boosting Organizational Productivity and Fraud Management”, International Journal of INTELLIGENT SYSTEMS AND APPLICATIONS IN ENGINEERING, vol. 11, no.10, pp. 1013–1023, 2023.
[41] Mohanarajesh Kommineni. (2023/6). Investigate Computational Intelligence Models Inspired By Natural Intelligence, Such As Evolutionary Algorithms And Artificial Neural Networks. Transactions On Latest Trends In Artificial Intelligence. 4. P30. Ijsdcs.
[42] Settibathini, V. S., Kothuru, S. K., Vadlamudi, A. K., Thammreddi, L., & Rangineni, S. (2023). Strategic analysis review of data analytics with the help of artificial intelligence. International Journal of Advances in Engineering Research, 26, 1-10.
[43] Sehrawat, S. K. (2023). The role of artificial intelligence in ERP automation: state-of-the-art and future directions. Trans Latest Trends Artif Intell, 4(4).
[44] Thallam, N. S. T. (2023). Comparative Analysis of Public Cloud Providers for Big Data Analytics: AWS, Azure, and Google Cloud. International Journal of AI, BigData, Computational and Management Studies, 4(3), 18-29.
[45] Mukkala, S. R. (2023). A Proficient Hospital Ratings Aware Patient Churn Prediction And Prevention System Using Abg-Fuzzy And Ner-Gfjdkmeans. Educational Administration: Theory and Practice, 29 (03), 1407-1424 Doi: 10.53555/kuey. v29i3, 9511.
[46] Rajesh Kumar Kanji, Vinodkumar Reddy Surasani, Naveen Kumar Kotha and Uday Kiran Chilakalapalli4 (2023). NLP-BASED INTER AND INTRA-SENTENCE RELATIONSHIP ANALYSIS-AWARE BANK CUSTOMER BEHAVIOR ANALYSIS AND PREFERENCE DETECTION USING GLSNSTM. Journal of Computational Analysis and Applications, 31(4), 1834-1857
[47] Varinder Kumar Sharma - Cloud-Edge Continuum in 5G: A Latency-Aware Network Design Review -International Scientific Journal of Engineering and Management Volume: 02 Issue: 03 | Mar – 2023. DOI: 10.55041/ISJEM00133
