Exploring the Systematic Use of LLMs for Microservices Generation. Trabelsi, I., Moha, N., & Gu�h�neuc, Y. In Beheshti, A., Benatallah, B., & Motahari, H., editors, Proceedings of the 5<sup>th</sup> International Workshop on AI-enabled Process Automation (AI-PA), pages 121–128, December, 2024. Springer. 8 pages.
Paper abstract bibtex The inflexibility of monolithic architectures results in a low software systems scalability and agility, compared to the modular systems based on microservices that can be independently developed and deployed. To facilitate the migration from monolithic to microservices architecture (MSA), researchers and practitioners have proposed various approaches. However, most of these works focus on automating the identification of microservices, and few have addressed the practical aspects of generating and deploying them. This gap leaves the migration process costly and complex, relying on expert intervention and manual labour to transform monolith systems into microservices. In this paper, we suggest that LLMs constitute an exciting but ill-studied means to automate key steps of the generation of microservices, including source code restructuring, API implementation, and integration of microservice patterns. We introduce (1) a systematic approach for generating LLM prompts to guide the migration process and (2) insights drawn from applying this approach to a real-world system. Our position challenges the current reliance on manual labour for MSA migration and supports LLM-assisted automation, encouraging discussions on the feasibility, benefits and wider impact of incorporating LLMs into the migration process. We introduce (1) a systematic approach for generating LLM prompts to guide the migration process and (2) insights drawn from applying this approach to a real-world system. Our position challenges the current reliance on manual labour for MSA migration and supports LLM-assisted automation, driving the discussion on the practicality, advantages and impact of including LLMs in the migration process.
@INPROCEEDINGS{Trabelsi24-AIPA-LLMsMicroservicesGeneration,
AUTHOR = {Imen Trabelsi and Naouel Moha and Yann-Ga�l Gu�h�neuc},
BOOKTITLE = {Proceedings of the 5<sup>th</sup> International Workshop on AI-enabled Process Automation (AI-PA)},
TITLE = {Exploring the Systematic Use of LLMs for Microservices
Generation},
YEAR = {2024},
OPTADDRESS = {},
OPTCROSSREF = {},
EDITOR = {Amin Beheshti and Boualem Benatallah and Hamid Motahari},
MONTH = {December},
NOTE = {8 pages.},
OPTNUMBER = {},
OPTORGANIZATION = {},
PAGES = {121–128},
PUBLISHER = {Springer},
OPTSERIES = {},
OPTVOLUME = {},
KEYWORDS = {Topic: <b>Evolution patterns</b>,
Rubrique : <b>patrons d'�volution</b>, Workshop: AI-PA},
URL = {http://www.ptidej.net/publications/documents/AIPA24.doc.pdf},
PDF = {http://www.ptidej.net/publications/documents/AIPA24.ppt.pdf},
ABSTRACT = {The inflexibility of monolithic architectures results in
a low software systems scalability and agility, compared to the
modular systems based on microservices that can be independently
developed and deployed. To facilitate the migration from monolithic
to microservices architecture (MSA), researchers and practitioners
have proposed various approaches. However, most of these works focus
on automating the identification of microservices, and few have
addressed the practical aspects of generating and deploying them.
This gap leaves the migration process costly and complex, relying on
expert intervention and manual labour to transform monolith systems
into microservices. In this paper, we suggest that LLMs constitute an
exciting but ill-studied means to automate key steps of the
generation of microservices, including source code restructuring, API
implementation, and integration of microservice patterns. We
introduce (1) a systematic approach for generating LLM prompts to
guide the migration process and (2) insights drawn from applying this
approach to a real-world system. Our position challenges the current
reliance on manual labour for MSA migration and supports LLM-assisted
automation, encouraging discussions on the feasibility, benefits and
wider impact of incorporating LLMs into the migration process. We
introduce (1) a systematic approach for generating LLM prompts to
guide the migration process and (2) insights drawn from applying this
approach to a real-world system. Our position challenges the current
reliance on manual labour for MSA migration and supports LLM-assisted
automation, driving the discussion on the practicality, advantages
and impact of including LLMs in the migration process.}
}
Downloads: 0
{"_id":"sorXh2kvt5GFYxKht","bibbaseid":"trabelsi-moha-guhneuc-exploringthesystematicuseofllmsformicroservicesgeneration-2024","author_short":["Trabelsi, I.","Moha, N.","Gu�h�neuc, Y."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Imen"],"propositions":[],"lastnames":["Trabelsi"],"suffixes":[]},{"firstnames":["Naouel"],"propositions":[],"lastnames":["Moha"],"suffixes":[]},{"firstnames":["Yann-Ga�l"],"propositions":[],"lastnames":["Gu�h�neuc"],"suffixes":[]}],"booktitle":"Proceedings of the 5<sup>th</sup> International Workshop on AI-enabled Process Automation (AI-PA)","title":"Exploring the Systematic Use of LLMs for Microservices Generation","year":"2024","optaddress":"","optcrossref":"","editor":[{"firstnames":["Amin"],"propositions":[],"lastnames":["Beheshti"],"suffixes":[]},{"firstnames":["Boualem"],"propositions":[],"lastnames":["Benatallah"],"suffixes":[]},{"firstnames":["Hamid"],"propositions":[],"lastnames":["Motahari"],"suffixes":[]}],"month":"December","note":"8 pages.","optnumber":"","optorganization":"","pages":"121–128","publisher":"Springer","optseries":"","optvolume":"","keywords":"Topic: <b>Evolution patterns</b>, Rubrique : <b>patrons d'�volution</b>, Workshop: AI-PA","url":"http://www.ptidej.net/publications/documents/AIPA24.doc.pdf","pdf":"http://www.ptidej.net/publications/documents/AIPA24.ppt.pdf","abstract":"The inflexibility of monolithic architectures results in a low software systems scalability and agility, compared to the modular systems based on microservices that can be independently developed and deployed. To facilitate the migration from monolithic to microservices architecture (MSA), researchers and practitioners have proposed various approaches. However, most of these works focus on automating the identification of microservices, and few have addressed the practical aspects of generating and deploying them. This gap leaves the migration process costly and complex, relying on expert intervention and manual labour to transform monolith systems into microservices. In this paper, we suggest that LLMs constitute an exciting but ill-studied means to automate key steps of the generation of microservices, including source code restructuring, API implementation, and integration of microservice patterns. We introduce (1) a systematic approach for generating LLM prompts to guide the migration process and (2) insights drawn from applying this approach to a real-world system. Our position challenges the current reliance on manual labour for MSA migration and supports LLM-assisted automation, encouraging discussions on the feasibility, benefits and wider impact of incorporating LLMs into the migration process. We introduce (1) a systematic approach for generating LLM prompts to guide the migration process and (2) insights drawn from applying this approach to a real-world system. Our position challenges the current reliance on manual labour for MSA migration and supports LLM-assisted automation, driving the discussion on the practicality, advantages and impact of including LLMs in the migration process.","bibtex":"@INPROCEEDINGS{Trabelsi24-AIPA-LLMsMicroservicesGeneration,\r\n AUTHOR = {Imen Trabelsi and Naouel Moha and Yann-Ga�l Gu�h�neuc},\r\n BOOKTITLE = {Proceedings of the 5<sup>th</sup> International Workshop on AI-enabled Process Automation (AI-PA)},\r\n TITLE = {Exploring the Systematic Use of LLMs for Microservices \r\n Generation},\r\n YEAR = {2024},\r\n OPTADDRESS = {},\r\n OPTCROSSREF = {},\r\n EDITOR = {Amin Beheshti and Boualem Benatallah and Hamid Motahari},\r\n MONTH = {December},\r\n NOTE = {8 pages.},\r\n OPTNUMBER = {},\r\n OPTORGANIZATION = {},\r\n PAGES = {121–128},\r\n PUBLISHER = {Springer},\r\n OPTSERIES = {},\r\n OPTVOLUME = {},\r\n KEYWORDS = {Topic: <b>Evolution patterns</b>, \r\n Rubrique : <b>patrons d'�volution</b>, Workshop: AI-PA},\r\n URL = {http://www.ptidej.net/publications/documents/AIPA24.doc.pdf},\r\n PDF = {http://www.ptidej.net/publications/documents/AIPA24.ppt.pdf},\r\n ABSTRACT = {The inflexibility of monolithic architectures results in \r\n a low software systems scalability and agility, compared to the \r\n modular systems based on microservices that can be independently \r\n developed and deployed. To facilitate the migration from monolithic \r\n to microservices architecture (MSA), researchers and practitioners \r\n have proposed various approaches. However, most of these works focus \r\n on automating the identification of microservices, and few have \r\n addressed the practical aspects of generating and deploying them. \r\n This gap leaves the migration process costly and complex, relying on \r\n expert intervention and manual labour to transform monolith systems \r\n into microservices. In this paper, we suggest that LLMs constitute an \r\n exciting but ill-studied means to automate key steps of the \r\n generation of microservices, including source code restructuring, API \r\n implementation, and integration of microservice patterns. We \r\n introduce (1) a systematic approach for generating LLM prompts to \r\n guide the migration process and (2) insights drawn from applying this \r\n approach to a real-world system. Our position challenges the current \r\n reliance on manual labour for MSA migration and supports LLM-assisted \r\n automation, encouraging discussions on the feasibility, benefits and \r\n wider impact of incorporating LLMs into the migration process. We \r\n introduce (1) a systematic approach for generating LLM prompts to \r\n guide the migration process and (2) insights drawn from applying this \r\n approach to a real-world system. Our position challenges the current \r\n reliance on manual labour for MSA migration and supports LLM-assisted \r\n automation, driving the discussion on the practicality, advantages \r\n and impact of including LLMs in the migration process.}\r\n}\r\n\r\n","author_short":["Trabelsi, I.","Moha, N.","Gu�h�neuc, Y."],"editor_short":["Beheshti, A.","Benatallah, B.","Motahari, H."],"key":"Trabelsi24-AIPA-LLMsMicroservicesGeneration","id":"Trabelsi24-AIPA-LLMsMicroservicesGeneration","bibbaseid":"trabelsi-moha-guhneuc-exploringthesystematicuseofllmsformicroservicesgeneration-2024","role":"author","urls":{"Paper":"http://www.ptidej.net/publications/documents/AIPA24.doc.pdf"},"keyword":["Topic: <b>Evolution patterns</b>","Rubrique : <b>patrons d'�volution</b>","Workshop: AI-PA"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"http://www.yann-gael.gueheneuc.net/Work/Publications/Biblio/complete-bibliography.bib","dataSources":["8vn5MSGYWB4fAx9Z4"],"keywords":["topic: <b>evolution patterns</b>","rubrique : <b>patrons d'�volution</b>","workshop: ai-pa"],"search_terms":["exploring","systematic","use","llms","microservices","generation","trabelsi","moha","gu�h�neuc"],"title":"Exploring the Systematic Use of LLMs for Microservices Generation","year":2024}