Clustered
^^ Clustered Placement Groups are for a set of EC2 instances that communicate a lot with each other and you want low network latency between them, so you want them in the same Availability Zone and as close together as possible (i.e. on the same rack).
^^^ Spread Placement Groups are for a set of EC2 instances that do not communicate much with eachother (i.e. low network latency is not an issue) but they each do critical workload that is a replication of each other, so you want them on differnt racks, possibly spread across multiple AZs. For Clustered Placement Groups think individual EC2s!
^^^^ Partitioned Placement groups, place EC2 instances into different partitions, each of these partitions is then in turn spread across multiple racks, possibly across multiple AZs. This is for HDFS, Cassandra usecases.
Clustered
^^ Clustered Placement Groups are for a set of EC2 instances that communicate a lot with each other and you want low network latency between them, so you want them in the same Availability Zone and as close together as possible (i.e. on the same rack).
^^^ Spread Placement Groups are for a set of EC2 instances that do not communicate much with eachother (i.e. low network latency is not an issue) but they each do critical workload that is a replication of each other, so you want them on differnt racks, possibly spread across multiple AZs. For Clustered Placement Groups think individual EC2s!
^^^^ Partitioned Placement groups, place EC2 instances into different partitions, each of these partitions is then in turn spread across multiple racks, possibly across multiple AZs. This is for HDFS, Cassandra usecases.
| status | not learned | measured difficulty | 37% [default] | last interval [days] | |||
|---|---|---|---|---|---|---|---|
| repetition number in this series | 0 | memorised on | scheduled repetition | ||||
| scheduled repetition interval | last repetition or drill |