Jiffy: Elastic Far-memory for Stateful Serverless Analytics

Abstract

Stateful serverless analytics can use a remote-memory system for inter-task communication and intermediate data, but existing systems allocate memory at job granularity for the job's entire lifetime. This causes low utilization or degraded performance when intermediate-data sizes vary during execution. Jiffy is an elastic far-memory system that meets a job's instantaneous memory demand at seconds timescales. It efficiently multiplexes capacity across concurrent jobs, reducing slower persistent-storage accesses and improving execution time by 1.6–2.5 times on production workloads. The implementation runs on Amazon EC2, supports multiple distributed programming models, and natively supports a broad class of analytics applications on AWS Lambda.

BibTeX

@inproceedings{jiffy,
  author = {Khandelwal, Anurag and Tang, Yupeng and Agarwal, Rachit and Akella, Aditya and Stoica, Ion},
  title = {{Jiffy: Elastic Far-memory for Stateful Serverless Analytics}},
  year = {2022},
  mon = {April},
  booktitle = {ACM EuroSys}
}