====== M ======

  * Maglev
    * "a large, distributed software system", the basis for the network load balancer in [[gcp:c|Cloud Load Balancing]].
    * https://research.google/pubs/pub44824/
  * MIG = Managed Instance Group
    * A group of identical instances sharing an instance template, and which can be resized
    * typically used with autoscaler but doesn't have to. Their unmanaged counterparts are only usable for load balancing, but not autocaling
    * regional or [[gcp:z|zonal]], with regional being generally recommended for zone failure resilience and simplicity
    * all instances in a group can be updated by specifying a new template in a [[gcp:r|rolling update]]
  * Monitoring
    * Per the [[https://sre.google|SRE books]]: collecting, processing, aggregating and displaying *real-time quantitative data* about a system, such as
      * query counts and types
      * error counts and types
      * processing times
      * server lifetimes