Data Platform Tips 64 - Recommended node configuration and VM sizes for HDInsight Clusters

The default and minimum recommended node configuration and virtual machine sizes for Azure HDInsight Clusters are listed below.

Size	vCPU	Memory: GiB	Temp storage (SSD) GiB	Max temp storage throughput: IOPS / Read MBps / Write MBps	Max data disks / throughput: IOPS	Max NICs / Expected network bandwidth (Mbps)
Standard_D3_v2	4	14	200	12000 / 187 / 93	16 / 16x500	4 / 3000
Standard_D4_v2	8	28	400	24000 / 375 / 187	32 / 32x500	8 / 6000
Standard_D5_v2	16	56	800	48000 / 750 / 375	64 / 64x500	8 / 12000
Standard_D12_v2	4	28	200	12000 / 187 / 93	16 / 16x500	4 / 3000
Standard_D13_v2	8	56	400	24000 / 375 / 187	32 / 32x500	8 / 6000
Standard_D14_v2	16	112	800	48000 / 750 / 375	64 / 64x500	8 / 12000
Standard_A1_v2	1	2	10	1000 / 20 / 10	2 / 2x500	2 / 250
Standard_A2_v2	2	4	20	2000 / 40 / 20	4 / 4x500	2 / 500
Standard_A4_v2	4	8	40	4000 / 80 / 40	8 / 8x500	4 / 1000

If you need more than 32 worker nodes in a cluster, select a head node size with at least 8 cores and 14 GB of RAM.

The only cluster types that have data disks are Kafka and HBase clusters with the Accelerated Writes feature enabled.