Object Storage System Settings
Unlike file systems, object storage is flat and stores objects in containers (OpenStack-Swift) or buckets (Amazon S3). Buckets and containers are conceptually the same and cannot contain other buckets or containers. As a result, you cannot create a container or bucket hierarchy. However, both Amazon S3 and OpenStack Swift support the concept of a “folder” for organization purposes in a manner that is conceptually similar to file systems. For Amazon S3 and OpenStack Swift, this can be accomplished by including a forward slash (/) character in the object name.
Specify how many containers or buckets you create and the number of objects per container or bucket. When you enable Pre-Test, WorkloadWisdom builds out the object storage system before the main workload run. The total number of objects created during Pre-Test is:
Total number of objects = (number of objects) x (number of buckets)
Note that the main workload that runs after Pre-Test completes may produce additional temporary objects for reasons described earlier in this section.
As with file workloads, you can specify a constant object size or create your own distribution of object sizes with custom bins. To change the size distribution method, select the option from the drop-down menu.
Object names can be auto-generated or user-defined.
Depending on the object storage system under test and the user privileges, you may or may not have the authorization to create Buckets or Containers on the fly, and you may have to use existing Buckets or Containers. If this is the case, you need to deselect Use unique bucket names and use user defined names for objects and buckets or containers so that the workload uses predictable bucket or container names per user.
For example, assume that the workload contains the following settings for the object storage system.
Then the following buckets are used to ensure unique Buckets are created and accessed per user.
If there are 3 workers for 3 user accounts then, according to screenshot, example should be:
If the Use unique bucket names option is checked:
Worker 1, user 1: bucket.0<1st unique number>, bucket.1<1st unique number>, …,bucket.9<1st unique number>
Worker 2, user 2: bucket.10<2nd unique number>, bucket.11<2nd unique number>, …,bucket.19<2nd unique number>
Worker 3, user 3: bucket.20<3rd unique number>, bucket.21<3rd unique number>, …,bucket.29<3rd unique number>
If the option Use unique bucket names is unchecked:
Worker 1, user 1: bucket.0, bucket.1, …,bucket.9
Worker 2, user 2: bucket.10, bucket.11, …,bucket.19
Worker 3, user 3: bucket.20, bucket.21, …,bucket.29
If there are 3 workers for 2 user accounts then, according to screenshot, example should be:
If the Use unique bucket names option is checked:
Worker 1, user 1: bucket.0<1st unique number>, bucket.1<1st unique number>, …,bucket.4<1st unique number>
Worker 2, user 2: bucket.10<2nd unique number>, bucket.11<2nd unique number>, …,bucket.19<2nd unique number>
Worker 3, user 1: bucket.5<1st unique number>, bucket.6<1st unique number>, …,bucket.9<1st unique number>
If the option Use unique bucket names is unchecked:
Worker 1, user 1: bucket.0, bucket.1, …,bucket.4
Worker 2, user 2: bucket.10, bucket.11, …,bucket.19
Worker 3, user 1: bucket.5, bucket.6, …,bucket.9
As in the file system section you can use a distribution of object sizes, with up to eight bins supported. Remove the default bins. Next, click +Add Bin. You can specify the size range by filling in the start (from) and to (end) sizes in bytes, kilobytes (KB), megabytes (MB) or gigabytes (GB). However, you cannot specify bins that are overlapping in object sizes. When this occurs, the “Add Bin” button appears grayed out and an error message displays when you mouse-over the Add Bin button.
You can change the ratio of size distributions by moving the sliders up to increase the value or down to decrease. You can set the slider maximum by selecting maximum percentage value the sliders can have in the top righthand corner of the bin distribution section.