Home > Storage > PowerScale (Isilon) > Product Documentation > Cloud > PowerScale: CloudPools and ECS > Performance
CloudPools is designed to move cold data from primary storage to the cloud. It is deliberately slow to ensure that it does not compete with things that are performance sensitive like SMB and NFS user activity. By default, CloudPools is using 10 threads per node which balances CloudPools CPU usage with other cluster functions. It is recommended to use the default number of threads for typical workloads. CloudPools does provide an option to modify the number of archive and recall threads. However, modifying the number of archive and recall threads can improve archive and recall performance but can also have significant impact on the CPU load of your system.
Here is as an example, the test is archiving and recalling 1 GB files between a four node Dell Isilon H500 cluster. The H500 cluster is connected using a physical server load balancer to a five node Dell ECS EX300 system. All equipment is connected on the same 10-Gigabit Ethernet network. With this configuration, it achieves ~ 261 MB/s (~ 65 MB/s per H500 node) for archive. It achieves ~702 MB/s (~ 175 MB/s per H500 node) for recall throughput of 1 GB files. The results are based on ideal test scenarios where there is no other load was on the systems or network. Typical environments would have additional loads on the systems, and so there will have different performance results.
Note: Contact your Dell representative if you want to configure higher number of threads.
CloudPools archive and recall performance are highly dependent upon many factors, such as the network bandwidth between the PowerScale cluster and the ECS system, available system resources and file size. These performance considerations would be: