375
views
1
recommends
+1 Recommend
0
shares
    • Review: found
    Is Open Access

    Review of 'Your data is your dogfood: DevOps in the astronomical observatory'

    Bookmark
    4
    Your data is your dogfood: DevOps in the astronomical observatory
    A good early example of the adoption of good practice in scientific computing infrastructures.
    Average rating:
        Rated 4 of 5.
    Level of importance:
        Rated 3 of 5.
    Level of validity:
        Rated 4 of 5.
    Level of completeness:
        Rated 4 of 5.
    Level of comprehensibility:
        Rated 4 of 5.
    Competing interests:
    none

    Reviewed article

    • Record: found
    • Abstract: found
    • Article: found
    Is Open Access

    Your data is your dogfood: DevOps in the astronomical observatory

    DevOps is the contemporary term for a software development culture that purposefully blurs distinction between software development and IT operations by treating "infrastructure as code." DevOps teams typically implement practices summarised by the colloquial directive to "eat your own dogfood;" meaning that software tools developed by a team should be used internally rather thrown over the fence to operations or users. We present a brief overview of how DevOps techniques bring proven software engineering practices to IT operations. We then discuss the application of these practices to astronomical observatories.
      Bookmark

      Review information

      Review text

      This article was written just over three years before this review and provides an interesting perspective onto some of  the thinking around developing and maintaining computing infrastructure for a large scientific project - the Large Synoptic Survey Telescope (LSST). This telescope was one of the first automated astronomical survey telescopes commissioned and the rate in which data was acquired was quite different to predecessors. It is quite appropriate, with hindsight, that the DevOps culture should take hold in this organisation, given the scale and complexity of it. 

      The authors prioritise the scientific data when discussing the culture of DevOps. This makes the article interesting to others down the line who are developing infrastructure for other instruments. Although not the first adoption of its kind, the article captures the spirit of a moment and stands the test of time. Some of the tools and practices may have changed slightly, but the need to apply Agile and Open techniques to research data acquisition and the platforms that enable it are as valid today as three years ago.

      Comments

      Comment on this review

      Version and Review History