GPU-Enabled Serverless Workflows for Efficient Multimedia Processing
Source abstract
Abstract
Serverless computing has introduced scalable event-driven processing in Cloud infrastructures. However, it is not trivial for multimedia processing to benefit from the elastic capabilities featured by serverless applications. To this aim, this paper introduces the evolution of a framework to support the execution of customized runtime environments in AWS Lambda in order to accommodate workloads that do not satisfy its strict computational requirements: increased execution times and the ability to use GPU-based resources. This has been achieved through the integration of AWS Batch, a managed service to deploy virtual elastic clusters for the execution of containerized jobs. In addition, a Functions Definition Language (FDL) is introduced for the description of data-driven workflows of functions. These workflows can simultaneously leverage both AWS Lambda for the highly-scalable execution of short jobs and AWS Batch, for the execution of compute-intensive jobs that can profit from GPU-based computing. To assess the developed open-source framework, we executed a case study for efficient serverless video processing. The workflow automatically generates subtitles based on the audio and applies GPU-based object recognition to the video frames, thus simultaneously harnessing different computing services. This allows for the creation of cost-effective highly-parallel scale-to-zero serverless workflows in AWS.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Scientific Computing and Data Management · Distributed and Parallel Computing Systems · Cloud Computing and Resource Management
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Instituto de Instrumentación para Imagen Molecular (I3M), Centro Mixto CSIC—Universitat Politècnica de València, Camino de Vera s/n, 46022 Valencia, Spain
Funding
Ministerio de Economía, Industria y Competitividad
References (38)
Show source-provided references
1. Amazon Web Services (2020, November 26). AWS Lambda. Available online: https://aws.amazon.com/lambda/.
2. Amazon Web Services (2020, November 26). Amazon Simple Storage Service (S3). Available online: https://aws.amazon.com/s3/.
3. Amazon Web Services (2020, November 26). Amazon API Gateway. Available online: https://aws.amazon.com/api-gateway/.
4. Serverless computing for container-based architectures10.1016/j.future.2018.01.022 ↗
5. Amazon Web Services (2020, November 26). AWS Batch. Available online: https://aws.amazon.com/batch/.
6. Jonas, E., Pu, Q., Venkataraman, S., Stoica, I., and Recht, B. (2017, January 25–27). Occupy the cloud: Distributed computing for the 99%. Proceedings of the 2017 Symposium on Cloud Computing—SoCC ’17, Santa Clara, CA, USA.10.1145/3127479.3128601 ↗
7. A framework and a performance assessment for serverless MapReduce on AWS Lambda10.1016/j.future.2019.02.057 ↗
8. Malawski, M., Gajek, A., Zima, A., Balis, B., and Figiela, K. (2017). Serverless execution of scientific workflows: Experiments with HyperFlow, AWS Lambda and Google Cloud Functions. Future Gener. Comput. Syst.
9. Jiang, Q., Lee, Y.C., and Zomaya, A.Y. (2017). Serverless Execution Of Scientific Workflows, Springer. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics).10.1007/978-3-319-69035-3_51 ↗
10. Skluzacek, T.J., Chard, R., Wong, R., Li, Z., Babuji, Y.N., Ward, L., Blaiszik, B., Chard, K., and Foster, I. (2019, January 9–13). Serverless Workflows for Indexing Large Scientific Data. Proceedings of the 5th International Workshop on Serverless Computing (WOSC ’19), Davis, CA, USA.10.1145/3366623.3368140 ↗
11. Chard, R., Skluzacek, T.J., Li, Z., Babuji, Y.N., Woodard, A., Blaiszik, B., Tuecke, S., Foster, I.T., and Chard, K. (2019). Serverless Supercomputing: High Performance Function as a Service for Science. CoRR.
12. Akkus, I.E., Chen, R., Rimac, I., Satzke, M.S.K., Beck, A., Aditya, P., and Hilt, V. (2018, January 11–13). SAND: Towards high-performance serverless computing. Proceedings of the 2018 USENIX Annual Technical Conference, USENIX ATC 2018, Boston, MA, USA.
13. Sethi, R.J., Gil, Y., Jo, H., and Philpot, A. (2013, January 21–25). Large-Scale Multimedia Content Analysis Using Scientific Workflows. Proceedings of the 21st ACM International Conference on Multimedia, Barcelona, Spain.10.1145/2502081.2502082 ↗
14. Gil, Y., Ratnakar, V., Kim, J., Gonzalez-Calero, P.A., Groth, P., Moody, J., and Deelman, E. (2011). Wings: Intelligent Workflow-Based Design of Computational Experiments. IEEE Intell. Syst., 26.10.1109/MIS.2010.9 ↗
15. Pegasus: A Workflow Management System for Science Automation10.1016/j.future.2014.10.008 ↗
16. Blockchain-based cloudlet management for multimedia workflow in mobile cloud computing10.1007/s11042-019-07900-x ↗
17. Zhang, M., Zhu, Y., Zhang, C., and Liu, J. (2019, January 21). Video processing with serverless computing. Proceedings of the 29th ACM Workshop on Network and Operating Systems Support for Digital Audio and Video—NOSSDAV ’19, Amherst, MA, USA.10.1145/3304112.3325608 ↗
18. Pérez, A., Caballer, M., Moltó, G., and Calatrava, A. (2019, January 8–12). A programming model and middleware for high throughput serverless computing applications. Proceedings of the ACM Symposium on Applied Computing, Limassol, Cyprus.10.1145/3297280.3297292 ↗
19. Amazon Web Services (2020, November 26). Amazon CloudWatch. Available online: https://aws.amazon.com/cloudwatch/.
20. Docker (2020, November 26). Docker Hub. Available online: https://hub.docker.com/.
21. Google Cloud (2020, November 26). Cloud Computing Services. Available online: https://cloud.google.com/.
22. Microsoft Azure (2020, November 26). Cloud Computing Services. Available online: https://azure.microsoft.com/en-us/.
23. Amazon Web Services (2020, November 26). Amazon EC2. Available online: https://aws.amazon.com/ec2/.
24. Amazon Web Services (2020, November 26). Amazon ECS. Available online: https://aws.amazon.com/ecs/.
25. NVIDIA (2020, December 02). NVIDIA Container Runtime. Available online: https://github.com/NVIDIA/nvidia-container-runtime.
26. Amazon Web Services (2020, November 26). AWS SDK for Python. Available online: https://aws.amazon.com/sdk-for-python/.
27. Enabling rootless Linux Containers in multi-user environments: The udocker tool10.1016/j.cpc.2018.05.021 ↗
28. Canonical (2020, December 02). Cloud-Init: The Standard for Customising Cloud Instances. Available online: https://cloud-init.io/.
29. FFmpeg (2020, November 26). FFmpeg—A Complete, Cross-Platform Solution to Record, Convert and Stream Audio and Video. Available online: https://www.ffmpeg.org/.
30. RunasSudo (2020, November 26). audio2srt. Available online: https://gitlab.com/RunasSudo/audio2srt.
31. Shmyrev, N. (2020, November 26). CMUSphinx Open Source Speech Recognition. Available online: http://cmusphinx.github.io/.
32. You only look once: Unified, real-time object detection
33. Redmon, J., and Farhadi, A. (2018). YOLOv3: An Incremental Improvement. CoRR.
34. Redmon, J. (2020, November 26). Darknet: Open Source Neural Networks in C. Available online: https://pjreddie.com/darknet/.
35. NVIDIA (2020, November 26). CUDA Zone. Available online: https://developer.nvidia.com/cuda-zone.
36. Agache, A., Brooker, M., Iordache, A., Liguori, A., Neugebauer, R., Piwonka, P., and Popa, D.M. (2020, January 2). Firecracker: Lightweight Virtualization for Serverless Applications. Proceedings of the 17th USENIX Symposium on Networked Systems Design and Implementation (NSDI 20), Santa Clara, CA, USA.
37. Casalboni, A. (2021, January 26). AWS Lambda Power Tuning. Available online: https://github.com/alexcasalboni/aws-lambda-power-tuning.
38. Towards European Open Science Commons: The EGI Open Data Platform and the EGI DataHub10.1016/j.procs.2016.08.294 ↗