research-article

On the Acceleration of FaaS Using Remote GPU Virtualization

Authors:

Diana M. Naranjo Delgado,

Manuel Contreras,

Germán Moltó,

Sebastián Risco,

Ignacio Blanquer,

Federico SillaAuthors Info & Claims

ICPE '23 Companion: Companion of the 2023 ACM/SPEC International Conference on Performance Engineering

Pages 157 - 164

https://doi.org/10.1145/3578245.3584933

Published: 15 April 2023 Publication History

Abstract

Serverless computing and, in particular, Function as a Service (FaaS) has introduced novel computational approaches with its highly-elastic capabilities, per-millisecond billing and scale-to-zero capacities, thus being of interest for the computing continuum. Services such as AWS Lambda allow efficient execution of event-driven short-lived bursty applications, even if there are limitations in terms of the amount of memory and the lack of GPU support for accelerated execution. To this aim, this paper analyses the suitability of including GPU support in AWS Lambda through the rCUDA middleware, which provides CUDA applications with remote GPU execution capabilities. A reference architecture for data-driven accelerated processing is introduced, based on elastic queues and event-driven object storage systems to manage resource contention and GPU scheduling. The benefits and limitations are assessed through a use case of sequence alignment. The results indicate that, for certain scenarios, the usage of remote GPUs in AWS Lambda represents a viable approach to reduce the execution time.

References

[1]

Amazon Web Services. [n.,d.] a. Amazon API Gateway. https://aws.amazon.com/api-gateway/

[2]

Amazon Web Services. [n.,d.] b. AWS Batch - Easy and Efficient Batch Computing Capabilities. https://aws.amazon.com/batch/

[3]

Amazon Web Services. [n.,d.] c. AWS Lambda. https://aws.amazon.com/lambda

[4]

Amazon Web Services. [n.,d.] d. Cloud Object Storage | Store & Retrieve Data Anywhere | Amazon Simple Storage Service (S3). https://aws.amazon.com/s3/

[5]

Microsoft Azure. [n.,d.]. Azure Functions. https://azure.microsoft.com/es-es/services/functions/#overview

[6]

Ioana Baldini, Paul Castro, Kerry Chang, Perry Cheng, Stephen Fink, Vatche Ishakian, Nick Mitchell, Vinod Muthusamy, Rodric Rabbah, Aleksander Slominski, and Philippe Suter. 2017. Serverless computing: Current trends and open problems. In Research Advances in Cloud Computing. Springer Singapore, Singapore, 1--20. https://doi.org/10.1007/978--981--10--5026--8_1 arxiv: 1706.03178

[7]

Alex Casalboni. [n.,d.]. AWS Lambda Power Tuning. https://github.com/alexcasalboni/aws-lambda-power-tuning

[8]

John Runwei Cheng and Mitsuo Gen. 2019. Accelerating genetic algorithms with GPU computing: A selective overview. Computers and Industrial Engineering, Vol. 128 (2019), 514--525. https://doi.org/10.1016/j.cie.2018.12.067

Digital Library

[9]

Angelos Christidis, Roy Davies, and Sotiris Moschoyiannis. 2019. Serving machine learning workloads in resource constrained environments: A serverless deployment example. Proceedings - 2019 IEEE 12th Conference on Service-Oriented Computing and Applications, SOCA 2019 (11 2019), 55--63. https://doi.org/10.1109/SOCA.2019.00016

[10]

Robert Cordingly, Navid Heydari, Hanfei Yu, Varik Hoang, Zohreh Sadeghi, and Wes Lloyd. 2021. Enhancing observability of serverless computing with the serverless application analytics framework. In ICPE 2021 - Companion of the ACM/SPEC International Conference on Performance Engineering. ACM, New York, NY, USA, 161--164. https://doi.org/10.1145/3447545.3451173

Digital Library

[11]

Philippe Despré s and Xun Jia. 2017. A review of GPU-based medical image reconstruction. Physica Medica, Vol. 42 (oct 2017), 76--92. https://doi.org/10.1016/j.ejmp.2017.07.024

[12]

Alex Ellis. [n.,d.]. OpenFaaS. https://www.openfaas.com/

[13]

Anshuj Garg, Purushottam Kulkarni, Umesh Bellur, and Sriram Yenamandra. 2021. FaaSter: Accelerated Functions-as-a-Service with Heterogeneous GPUs. In 2021 IEEE 28th International Conference on High Performance Computing, Data, and Analytics (HiPC). IEEE, 406--411. https://doi.org/10.1109/HiPC53243.2021.00057

[14]

Google Cloud. [n.,d.]. Cloud Functions - Event-driven Serverless Computing. https://cloud.google.com/functions/

[15]

Iguazio. [n.,d.]. Nuclio. https://nuclio.io/

[16]

Sergio Iserte, Javier Prades, Carlos Reaño, and Federico Silla. 2016. Increasing the Performance of Data Centers by Combining Remote GPU Virtualization with Slurm. In 2016 16th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid). 98--101.

Digital Library

[17]

Zhipeng Jia and Emmett Witchel. 2021. Nightcore: Efficient and scalable serverless computing for latency-sensitive, interactive microservices. In International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS, Vol. 15. ACM, New York, NY, USA, 152--166. https://doi.org/10.1145/3445814.3446701

Digital Library

[18]

Jaewook Kim, Tae Joon Jun, Daeyoun Kang, Dohyeun Kim, and Daeyoung Kim. 2018. GPU Enabled Serverless Computing Framework. In Proceedings - 26th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, PDP 2018. Institute of Electrical and Electronics Engineers Inc., 533--540. https://doi.org/10.1109/PDP2018.2018.00090

[19]

Microsoft. [n.,d.]. Azure AKS. https://azure.microsoft.com/services/kubernetes-service/#overview

[20]

Diana M. Naranjo, Sebastiá n Risco, Carlos de Alfonso, Alfonso Pé rez, Ignacio Blanquer, and Germá n Moltó. 2020. Accelerated serverless computing based on GPU virtualization. J. Parallel and Distrib. Comput., Vol. 139 (may 2020), 32--42. https://doi.org/10.1016/J.JPDC.2020.01.004

Digital Library

[21]

National Library of Medicine. [n.,d.]. BLAST: Basic Local Alignment Search Tool. https://blast.ncbi.nlm.nih.gov/Blast.cgi

[22]

Alfonso Pé rez, Germá n Moltó, Miguel Caballer, and Amanda Calatrava. 2018. Serverless computing for container-based architectures. Future Generation Computer Systems, Vol. 83 (jun 2018), 50--59. https://doi.org/10.1016/j.future.2018.01.022

[23]

Alfonso Perez, Sebastian Risco, Diana Maria Naranjo, Miguel Caballer, and Germán Moltó. 2019. On-premises serverless computing for event-driven data processing applications. In IEEE International Conference on Cloud Computing, CLOUD, Vol. 2019-July. 414--421. https://doi.org/10.1109/CLOUD.2019.00073

[24]

Javier Prades, Carlos Rea n o, and Federico Silla. 2019. On the effect of using rCUDA to provide CUDA acceleration to Xen virtual machines. Cluster Computing, Vol. 22, 1 (2019), 185--204. https://doi.org/10.1007/s10586-018--2845-0

Digital Library

[25]

Javier Prades and Federico Silla. 2018. Made-to-Measure GPUs on Virtual Machines with rCUDA. In The 47th International Conference on Parallel Processing, ICPP 2018, Workshop Proceedings, Eugene, OR, USA, August 13--16, 2018. ACM, 19:1--19:8. https://doi.org/10.1145/3229710.3229741

Digital Library

[26]

Chandra Prakash, Anshuj Garg, Umesh Bellur, Purushottam Kulkarni, Uday Kurkure, Hari Sivaraman, and Lan Vu. 2021. Optimizing Goodput of Real-time Serverless Functions using Dynamic Slicing with vGPUs. In Proceedings - 2021 IEEE International Conference on Cloud Engineering, IC2E 2021. Institute of Electrical and Electronics Engineers Inc., 60--70. https://doi.org/10.1109/IC2E52221.2021.00020

[27]

Carlos Reano and Federico Silla. 2015. A Performance Comparison of CUDA Remote GPU Virtualization Frameworks. In 2015 IEEE International Conference on Cluster Computing. IEEE, 488--489. https://doi.org/10.1109/CLUSTER.2015.76

Digital Library

[28]

Carlos Rea n o, Federico Silla, Gilad Shainer, and Scot Schultz. 2015. Local and Remote GPUs Perform Similar with EDR 100G InfiniBand. In Proceedings of the Industrial Track of the 16th International Middleware Conference on ZZZ - Middleware Industry '15. ACM Press, New York, New York, USA, 1--7. https://doi.org/10.1145/2830013.2830015

Digital Library

[29]

Sebastiá n Risco and Germá n Moltó. 2021. GPU-Enabled Serverless Workflows for Efficient Multimedia Processing. Applied Sciences, Vol. 11, 4 (feb 2021), 1438. https://doi.org/10.3390/app11041438

[30]

Sebastiá n Risco, Germá n Moltó, Diana M. Naranjo, and Ignacio Blanquer. 2021. Serverless Workflows for Containerised Applications in the Cloud Continuum. Journal of Grid Computing, Vol. 19, 3 (sep 2021), 30. https://doi.org/10.1007/s10723-021-09570--2

Digital Library

[31]

Klaus Satzke, Istemi Ekin Akkus, Ruichuan Chen, Ivica Rimac, Manuel Stein, Andre Beck, Paarijaat Aditya, Manohar Vanga, and Volker Hilt. 2021. Efficient GPU Sharing for Serverless Workflows. In HiPS 2021 - Proceedings of the 1st Workshop on High Performance Serverless Computing, co-located with HPDC 2021. ACM, New York, NY, USA, 17--24. https://doi.org/10.1145/3452413.3464785

Digital Library

[32]

Amazon Web Services. [n.,d.] a. Amazon ECR. https://aws.amazon.com/ecr/

[33]

Amazon Web Services. [n.,d.] b. Amazon EFS. https://aws.amazon.com/efs/

[34]

Amazon Web Services. [n.,d.] c. Amazon EKS. https://aws.amazon.com/eks/

[35]

Amazon Web Services. [n.,d.] d. Amazon SQS. https://aws.amazon.com/sqs/

[36]

Amazon Web Services. 2020. AWS Lambda now supports up to 10 GB of memory and 6 vCPU cores for Lambda Functions. https://aws.amazon.com/about-aws/whats-new/2020/12/aws-lambda-supports-10gb-memory-6-vcpu-cores-lambda-functions/

[37]

Vaishaal Shankar, Karl Krauth, Kailas Vodrahalli, Qifan Pu, Benjamin Recht, Ion Stoica, Jonathan Ragan-Kelley, Eric Jonas, and Shivaram Venkataraman. 2020. Serverless linear algebra. In SoCC 2020 - Proceedings of the 2020 ACM Symposium on Cloud Computing, Vol. 15. Association for Computing Machinery, Inc, New York, NY, USA, 281--295. https://doi.org/10.1145/3419111.3421287

Digital Library

[38]

Federico Silla, Sergio Iserte, Carlos Reaño, and Javier Prades. 2017. On the benefits of the remote GPU virtualization mechanism: The rCUDA case. Concurrency and Computation: Practice and Experience, Vol. 29, 13 (2017), e4072. e4072 cpe.4072.

[39]

Sijun Tan, Brian Knott, Yuan Tian, and David J. Wu. 2021. CryptGPU: Fast privacy-preserving machine learning on the GPU. In Proceedings - IEEE Symposium on Security and Privacy, Vol. 2021-May. Institute of Electrical and Electronics Engineers Inc., 1021--1038. https://doi.org/10.1109/SP40001.2021.00098 arxiv: 2104.10949

[40]

Manuel Ujaldón. 2016. CUDA achievements and GPU challenges ahead. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 9756. 207--217. https://doi.org/10.1007/978--3--319--41778--3_20

Cited By

Teguia DChen JBitchebe SBalmau OTchana ASchiavoni VEdinger JCao JJin Z(2024)vPIM: Processing-in-Memory VirtualizationProceedings of the 25th International Middleware Conference10.1145/3652892.3700782(417-430)Online publication date: 2-Dec-2024
https://dl.acm.org/doi/10.1145/3652892.3700782

Index Terms

On the Acceleration of FaaS Using Remote GPU Virtualization
1. Computing methodologies
  1. Distributed computing methodologies

Recommendations

The SPEC cloud group's research vision on FaaS and serverless architectures
WoSC '17: Proceedings of the 2nd International Workshop on Serverless Computing

Cloud computing enables an entire ecosystem of developing, composing, and providing IT services. An emerging class of cloud-based software architectures, serverless, focuses on providing software architects the ability to execute arbitrary functions ...
FaaS for IoT: Evolving Serverless towards Deviceless in I/Oclouds
Abstract
The burgeoning paradigms of Fog and Edge computing propose delegating Cloud-related tasks to the network’s periphery, thus placing computational resources closer to data producers. This shift promises to boost the performance of IoT-based ...
Highlights
- Deployment of FaaS-based applications on a fleet of IoTs.
- A tool to create applications as workflows exploiting FaaS paradigm via I/OCloud.
- Functions configuration without direct devices interaction mediated via a dashboard.
- ...
SeBS: a serverless benchmark suite for function-as-a-service computing
Middleware '21: Proceedings of the 22nd International Middleware Conference

Function-as-a-Service (FaaS) is one of the most promising directions for the future of cloud services, and serverless functions have immediately become a new middleware for building scalable and cost-efficient microservices and appli cations. However, ...

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences

ICPE '23 Companion: Companion of the 2023 ACM/SPEC International Conference on Performance Engineering

April 2023

421 pages

ISBN:9798400700729

DOI:10.1145/3578245

General Chairs:
Marco Vieira
University of Coimbra, Portugal
,
Valeria Cardellini
University of Rome Tor Vergata, Italy
,
Program Chairs:
Antinisca Di Marco
University of L'Aquila, Italy
,
Petr Tuma
Charles University, Czechia

Copyright © 2023 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 15 April 2023

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Funding Sources

European Commission
MCIN/AEI/10.13039/501100011033

Conference

ICPE '23

Sponsor:

ICPE '23: ACM/SPEC International Conference on Performance Engineering

April 15 - 19, 2023

Coimbra, Portugal

Acceptance Rates

Overall Acceptance Rate 252 of 851 submissions, 30%

Upcoming Conference

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

1
Total Citations
View Citations
157
Total Downloads

Downloads (Last 12 months)50
Downloads (Last 6 weeks)8

Reflects downloads up to 15 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

Teguia DChen JBitchebe SBalmau OTchana ASchiavoni VEdinger JCao JJin Z(2024)vPIM: Processing-in-Memory VirtualizationProceedings of the 25th International Middleware Conference10.1145/3652892.3700782(417-430)Online publication date: 2-Dec-2024
https://dl.acm.org/doi/10.1145/3652892.3700782

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Media

Figures

Other

Tables

View Table of Contents