Putting in all the stops: execution control for JavaScript

Baxter, S.; Nigam, Rachit; Politz, Joe Gibbs; Krishnamurthi, Shriram; Guha, Arjun

doi:10.1145/3192366.3192370

Cited by 8 publications

(12 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Checkpoints in Kappa are implemented using continuations [74], a language-level mechanism executed in user mode. Continuations are a well understood technique for suspending and resuming execution, and have been used in past systems for checkpointing and fault tolerance [79,88], task migration [28,78,79,88], asynchronous calls [51,54], context switching [86], and debugging [16]. Kappa's checkpointing technique is nearly identical to that of previous systems, although there might be implementation differences.…”

Section: Checkpointingmentioning

confidence: 99%

“…Our implementation is specialized to Python, and for simplicity we describe our methods in this context. However, continuations have been implemented in many other languages and platforms including Java [28,78,79,86,88], JavaScript [16,51,54], and .NET [66], and our techniques can be easily extended to other languages.…”

Section: Checkpointingmentioning

confidence: 99%

“…We implemented four SQL queries from the TPC-DS benchmark [90] 15 as multi-stage MapReduce jobs on top of Kappa tasks. Tasks in each stage are launched using map_spawn, data is shuffled using Kappa queues 16 , and query logic is implemented using the pandas library [53].…”

Section: Applicationsmentioning

confidence: 99%

“…Each worker 15 We used the queries for which PyWren implementations are available [69] (Q1, Q16, Q94, and Q95); we modified them to improve their performance. 16 These queues are configured to pass data through Redis. By the fault tolerance setup ( § 5), Kappa stores two copies of all shuffle data.…”

Section: Applicationsmentioning

confidence: 99%

“…These limitations are not fundamental-prior work has shown how to implement these features in continuation passing style, a paradigm that our approach depends on. For example, Stopify [16] shows how to generate continuations for nested functions and exception handling in JavaScript. Other restrictions on input code.…”

Section: Limitations and Future Workmentioning

confidence: 99%

See 4 more Smart Citations

Kappa

Zhang

Fang

Panda³

et al. 2020

Proceedings of the 11th ACM Symposium on Cloud Computing

View full text Add to dashboard Cite

Serverless computing has recently emerged as a new paradigm for running software on the cloud. In this paradigm, programs need to be expressed as a set of short-lived tasks, each of which can complete within a short bounded time (e.g., 15 minutes on AWS Lambda). Serverless computing is beneficial to cloud providers-by allowing them to better utilize resources-and to users-by simplifying management and enabling greater elasticity. However, developing applications to run in this environment is challenging, requiring users to appropriately partition their code, develop new coordination mechanisms, and deal with failure recovery. In this paper, we propose Kappa, a framework that simplifies serverless development. It uses checkpointing to handle lambda function timeouts, and provides concurrency mechanisms that enable parallel computation and coordination. CCS Concepts • Computer systems organization → Cloud computing.

show abstract