My Favorite Data Platform Feature

  • Not only is provisioning databases in Azure quick and simple, but billing and scaling in terms of DTU and RU/s is such a game changer that organization who are still entirely on-prem just havn't had a chance to wrap their heads around it yet.

    Another little known Azure billing feature are 'Spot Virtual Machines' which leverage unused compute capacity at up to a 90% discount. This can be an option for development or even production batch processing VMs that are idle most of the time and don't require a high SLA.

    https://azure.microsoft.com/en-us/services/virtual-machines/spot/#features

    "Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho

  • Http verb-specific route handlers in .NET 6 filled the last missing piece of the data access puzzle

    Aus dem Paradies, das Cantor uns geschaffen, soll uns niemand vertreiben können

  • I really like Vertica's export to Parquet feature and also the incredible speed at which it can import Parquet files directly from an AWS S3 bucket.

    It's also no slouch at importing JSON attributes from files directly into columns provided those column names reflect the JSON path for the attribute.

    Under the hood it uses Apache Arrow for translating its columnar format to other columnar formats.

    It uses a C library called YAJL to import JSON.

    I'm very impressed with Snowflake and the way it uses integrations and stages.

    It has its own task scheduler that you setup using a SQL CREATE TASK command. In fact everything seems to have a well thought out SQL command to support it.

    The thing that is most impressive is how comprehensive the SQL implementation is, especially considering how relatively young Snowflake is

Viewing 4 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic. Login to reply