Articles producció científica> Enginyeria Informàtica i Matemàtiques

MLLESS: Achieving cost efficiency in serverless machine learning training

  • Identification data

    Identifier: imarina:9330487
    Authors:
    Sarroca, PGSánchez-Artigas, M
    Abstract:
    Function-as-a-Service (FaaS) has raised a growing interest in how to “tame” serverless computing to enable domain-specific use cases such as data-intensive applications and machine learning (ML), to name a few. Recently, several systems have been implemented for training ML models. Certainly, these research articles are significant steps in the correct direction. However, they do not completely answer the nagging question of when serverless ML training can be more cost-effective compared to traditional “serverful” computing. To help in this endeavor, we propose MLLESS, a FaaS-based ML training prototype built atop IBM Cloud Functions. To boost cost-efficiency, MLLESS implements two innovative optimizations tailored to the traits of serverless computing: on one hand, a significance filter, to make indirect communication more effective, and on the other hand, a scale-in auto-tuner, to reduce cost by benefiting from the FaaS sub-second billing model (often per 100 ms). Our results certify that MLLESS can be 15X faster than serverful ML systems [27] at a lower cost for sparse ML models that exhibit fast convergence such as sparse logistic regression and matrix factorization. Furthermore, our results show that MLLESS can easily scale out to increasingly large fleets of serverless workers.
  • Others:

    Author, as appears in the article.: Sarroca, PG; Sánchez-Artigas, M
    Department: Enginyeria Informàtica i Matemàtiques
    URV's Author/s: Sanchez Artigas, Marc
    Keywords: Serverless computing Machine learning Function-as-a-service
    Abstract: Function-as-a-Service (FaaS) has raised a growing interest in how to “tame” serverless computing to enable domain-specific use cases such as data-intensive applications and machine learning (ML), to name a few. Recently, several systems have been implemented for training ML models. Certainly, these research articles are significant steps in the correct direction. However, they do not completely answer the nagging question of when serverless ML training can be more cost-effective compared to traditional “serverful” computing. To help in this endeavor, we propose MLLESS, a FaaS-based ML training prototype built atop IBM Cloud Functions. To boost cost-efficiency, MLLESS implements two innovative optimizations tailored to the traits of serverless computing: on one hand, a significance filter, to make indirect communication more effective, and on the other hand, a scale-in auto-tuner, to reduce cost by benefiting from the FaaS sub-second billing model (often per 100 ms). Our results certify that MLLESS can be 15X faster than serverful ML systems [27] at a lower cost for sparse ML models that exhibit fast convergence such as sparse logistic regression and matrix factorization. Furthermore, our results show that MLLESS can easily scale out to increasingly large fleets of serverless workers.
    Thematic Areas: Theoretical computer science Software Matemática / probabilidade e estatística Interdisciplinar Hardware and architecture Engenharias iv Engenharias iii Computer science, theory & methods Computer networks and communications Ciência da computação Artificial intelligence
    licence for use: https://creativecommons.org/licenses/by/3.0/es/
    Author's mail: marc.sanchez@urv.cat
    Author identifier: 0000-0002-9700-7318
    Record's date: 2024-08-03
    Papper version: info:eu-repo/semantics/publishedVersion
    Link to the original source: https://www.sciencedirect.com/science/article/pii/S074373152300134X
    Licence document URL: https://repositori.urv.cat/ca/proteccio-de-dades/
    Papper original source: Journal Of Parallel And Distributed Computing. 183
    APA: Sarroca, PG; Sánchez-Artigas, M (2024). MLLESS: Achieving cost efficiency in serverless machine learning training. Journal Of Parallel And Distributed Computing, 183(), -. DOI: 10.1016/j.jpdc.2023.104764
    Article's DOI: 10.1016/j.jpdc.2023.104764
    Entity: Universitat Rovira i Virgili
    Journal publication year: 2024
    Publication Type: Journal Publications
  • Keywords:

    Artificial Intelligence,Computer Networks and Communications,Computer Science, Theory & Methods,Hardware and Architecture,Software,Theoretical Computer Science
    Serverless computing
    Machine learning
    Function-as-a-service
    Theoretical computer science
    Software
    Matemática / probabilidade e estatística
    Interdisciplinar
    Hardware and architecture
    Engenharias iv
    Engenharias iii
    Computer science, theory & methods
    Computer networks and communications
    Ciência da computação
    Artificial intelligence
  • Documents:

  • Cerca a google

    Search to google scholar