Development and Deployment at Facebook

Feitelson, Dror G.; Frachtenberg, Eitan; Beck, Kent

doi:10.1109/mic.2013.25

Cited by 216 publications

(177 citation statements)

References 7 publications

Supporting

Mentioning

174

Contrasting

Unclassified

Order By: Relevance

“…At companies like Etsy, features are rolled out as soon as they are deemed ready, independently of a defined release plan. Facebook claims to occupy a middle ground between release trains and strict CD, where most features are rolled out into production the same day they are ready, while a fraction of particularly sensitive changes (e.g., those related to privacy) are rolled out once a week [14]. In practice, these models result in frequent, but tiny, changes to the production environment, in the most extreme cases multiple times a day.…”

Section: Continuous Deliverymentioning

confidence: 99%

“…This practice increases both, the business agility of the application provider, as well as the likelihood of releasing badlyperforming changes to production [42]. A consequence is that SaaS applications with such release models tend to be in a state of perpetual development [14] -there is never a stable, released and tagged version which is not supposed to be touched by developers.…”

Section: Continuous Deliverymentioning

confidence: 99%

See 1 more Smart Citation

Runtime metric meets developer - Building better cloud applications using feedback

Cito

Leitner

Gall

et al. 2015

Preprint

View full text Add to dashboard Cite

A unifying theme of many ongoing trends in software engineering is a blurring of the boundaries between building and operating software products. In this paper, we explore what we consider to be the logical next step in this succession: integrating runtime monitoring data from production deployments of the software into the tools developers utilize in their daily workflows (i.e., IDEs) to enable tighter feedback loops. We refer to this notion as feedback-driven development (FDD).This more abstract FDD concept can be instantiated in various ways, ranging from IDE plugins that implement feedback-driven refactoring and code optimization to plugins that predict performance and cost implications of code changes prior to even deploying the new version of the software. We demonstrate existing proof-of-concept realizations of these ideas and illustrate our vision of the future of FDD and cloud-based software development in general. Further, we discuss the major challenges that need to be solved before FDD can achieve mainstream adoption.

show abstract

Section: Continuous Deliverymentioning

confidence: 99%

Section: Continuous Deliverymentioning

confidence: 99%

Runtime metric meets developer - Building better cloud applications using feedback

Cito

Leitner

Gall

et al. 2015

Preprint

View full text Add to dashboard Cite

show abstract

“…As a starting point, we studied [12,15,27], which we considered standard CD literature at the time we conducted our pre-study (the mapping study described in [28], which we also consider seminal for the field, was not yet available). From those sources, we extracted practices and categorized them into five categories: automation, rollout, quality assurance, issue detection and handling, and awareness.…”

Section: Pre-studymentioning

confidence: 99%

“…To reduce the burden of long-living parallel development branches, companies have started to adopt the idea of trunk-based development [12], wherein all teams contribute to a single branch, usually called master, trunk, or mainline. Trunk-based development requires means to "switch on or off" individual code, if it is not ready for production.…”

Section: Continuous Integrationmentioning

confidence: 99%

See 1 more Smart Citation

An empirical study on principles and practices of continuous delivery and deployment

Schermann

Cito

Leitner

et al. 2016

Preprint

View full text Add to dashboard Cite

Despite substantial recent research activity related to continuous delivery and deployment (CD), there has not yet been a systematic, empirical study on how the practices often associated with continuous deployment have found their way into the broader software industry. This raises the question to what extent our knowledge of the area is dominated by the peculiarities of a small number of industrial leaders, such as Facebook. To address this issue, we conducted a mixed-method empirical study, consisting of a pre-study on literature, qualitative interviews with 20 software developers or release engineers with heterogeneous backgrounds, and a Web-based quantitative survey that attracted 187 complete responses. A major trend in the results of our study is that architectural issues are currently one of the main barriers for CD adoption. Further, feature toggles as an implementation technique for partial rollouts lead to unwanted complexity, and require research on better abstractions and modelling techniques for runtime variability. Finally, we conclude that practitioners are in need for more principled approaches to release decision making, e.g., which features to conduct A/B tests on, or which metrics to evaluate.

show abstract

A qualitative study of DevOps usage in practice

Erich¹,

Amrit²,

Daneva

2017

J Software Evolu Process

150

112

View full text Add to dashboard Cite

Organizations are introducing agile and lean software development techniques in operations to increase the pace of their software development process and to improve the quality of their software. They use the term DevOps, a portmanteau of development and operations, as an umbrella term to describe their efforts. In this paper, we describe the ways in which organizations implement DevOps and the outcomes they experience. We first summarize the results of a systematic literature review that we performed to discover what researchers have written about DevOps. We then describe the results of an exploratory interview‐based study involving 6 organizations of various sizes that are active in various industries. As part of our findings, we observed that all organizations were positive about their experiences and only minor problems were encountered while adopting DevOps.

show abstract

Development and Deployment at Facebook

Cited by 216 publications

References 7 publications

Runtime metric meets developer - Building better cloud applications using feedback

Runtime metric meets developer - Building better cloud applications using feedback

An empirical study on principles and practices of continuous delivery and deployment

A qualitative study of DevOps usage in practice

Contact Info

Product

Resources

About