Home > Storage > PowerScale (Isilon) > Industry Solutions and Verticals > Hadoop and Big Data > Dell PowerScale and Cloudera CDP Private Cloud Base Reference Architecture > Sizing recommendation for storage allocation
The following table provides recommendations for storage allocation.
Node and role | Disk layout | Description |
Management or master | 2 x 500 GB OS (RAID 1) Swap partition <= 2 GB 4 x 500 GB RAID 10 (database) 1 x 500 GB RAID 0 - ZooKeeper | Avoid fracturing the file system layout into multiple smaller file systems. Instead, keep a separate ‘/’ and /var. |
Compute nodes | 2 x 500 GB OS (RAID 1) Approximately 20% of total DFS storage (in this case, PowerScale storage) must be provisioned as intermediate storage on these nodes. The storage can be direct-attached SAS or SATA drives, or a pair of SSD drives of sufficient capacity. Distribute the 20% of capacity evenly across all the NodeManager nodes, with its own mount-point and file system. | Avoid fracturing the file system layout into multiple smaller file systems. Instead, keep a separate ‘/’ and /var. For example, for 10 TB of total storage in PowerScale, 2 TB is needed for intermediate storage. Having more or faster local spindles will speed up the intermediate shuffle stage of MapReduce. |