Serverless Applications: Why, When, and How?

Eismann, Simon; Scheuner, Joel; Eyk, Erwin van; Schwinger, Maximilian; Grohmann, Johannes; Herbst, Nikolas; Abad, Cristina L.; Iosup, Alexandru

doi:10.1109/ms.2020.3023302

Cited by 82 publications

(42 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other works investigate the efficiency of serverless systems for different classes workloads, namely ML training [37], latency-critical microservices [38], data-intensive applications [39]- [41], and confidential computations [42]. Eismann et al categorizes open-source serverless applications according to their non-performance characteristics [43]. Shahrad et al analyzes invocation frequency and execution time distributions of applications in Azure Functions and explores the design space of function instance keep-alive policies [16].…”

Section: Related Workmentioning

confidence: 99%

Analyzing Tail Latency in Serverless Clouds with STeLLAR

Ustiugov

Amariucai

Grot

2021

2021 IEEE International Symposium on Workload Characterization (IISWC)

View full text Add to dashboard Cite

Serverless computing has seen rapid adoption because of its instant scalability, flexible billing model, and economies of scale. In serverless, developers structure their applications as a collection of functions invoked by various events like clicks, and cloud providers take responsibility for cloud infrastructure management. As with other cloud services, serverless deployments require responsiveness and performance predictability manifested through low average and tail latencies. While the average end-to-end latency has been extensively studied in prior works, existing papers lack a detailed characterization of the effects of tail latency in real-world serverless scenarios and their root causes.In response, we introduce STeLLAR, an open-source serverless benchmarking framework, which enables an accurate performance characterization of serverless deployments. STeLLAR is provider-agnostic and highly configurable, allowing the analysis of both end-to-end and per-component performance with minimal instrumentation effort. Using STeLLAR, we study three leading serverless clouds and reveal that storage accesses and bursty function invocation traffic are key factors impacting tail latency in modern serverless systems. Finally, we identify important factors that do not contribute to latency variability, such as the choice of language runtime.

show abstract

Section: Related Workmentioning

confidence: 99%

Analyzing Tail Latency in Serverless Clouds with STeLLAR

Ustiugov

Amariucai

Grot

2021

2021 IEEE International Symposium on Workload Characterization (IISWC)

View full text Add to dashboard Cite

show abstract

“…The cost effectiveness of FaaS for low-rate and bursty computations has been observed and reported by various studies [42,66,67] and has been leveraged by frameworks like LI-BRA [103] and Spock [76]. This unique FaaS economic model has revenue and performance implications for both the cloud provider and the application developer.…”

Section: Serverless Economic Modelmentioning

confidence: 98%

SoK: Function-As-A-Service: From An Application Developer’s Perspective

Raza¹,

Matta²,

Akhtar³

et al. 2021

Journal of Systems Research

View full text Add to dashboard Cite

Serverless computing is a fast-moving, highly active research area. As a growing number of practitioners start using serverless and new researchers explore advancing this domain, well-written systemization of knowledge (SoK) articles like this paper are priceless to the community. The authors of this paper have done a great job providing clear categorization and insightful descriptions of various aspects of serverless systems as perceived by developers. At parts, the way the authors connected different dots provides new insights even to the experts in the field. I am glad to see the publication of such a valuable contribution as the inaugurating article in the Serverless Area of Journal of Systems Research.

show abstract

“…Server-less computing is any computing platform that hides server usage from developers and runs code on-demand automatically scaled and billed only for the time the code is running [4]. Based on Eismann, et al [5] analysis of 89 server-less applications, they find that the most commonly reported reasons for the adoption of server-less are to save costs for irregular or bursty workloads, to avoid operational concerns, and for the built-in scalability. Serverless applications are most commonly used for short-running tasks with low data volume and bursty workloads but are also frequently used for latency-critical, high-volume core functionality.…”

Section: State Of the Art In Software Architecturementioning

confidence: 99%

The impact of the software architecture on the developer productivity

Kokrehel

Bilicki

2022

Pollack

View full text Add to dashboard Cite

Distinct technological trends seriously influence the modern software architectures. In this paper, four different software architectures and framework combinations were generally compared. The basis for the analysis is the developer's productivity. In a 3 year-long research and development project, a real-world telemedicine application was efficiently implemented four times with various software architectures and architectural patterns. More than 5,000 person-hours were spent on carrying out them. At present, a unique dataset is available, which provides the opportunity to compare the cost of spent person-hours in different approaches. The goal of this research is to describe the measurement approach, the dataset and the applied architectures considering the software developer's productivity.

show abstract

Serverless Applications: Why, When, and How?

Cited by 82 publications

References 8 publications

Analyzing Tail Latency in Serverless Clouds with STeLLAR

Analyzing Tail Latency in Serverless Clouds with STeLLAR

SoK: Function-As-A-Service: From An Application Developer’s Perspective

The impact of the software architecture on the developer productivity

Contact Info

Product

Resources

About