sidebar | permalink | keywords | summary |
---|---|---|---|
sidebar |
concepts/metrics.html |
metrics, GiB, IOPS, latency, durability |
Metrics calculations and definitions of Keystone Services |
The following terms and definitions are used within the NetApp Keystone (Keystone) STaaS service to measure metrics:
-
Capacity measurement units: GiB, TiB, and PiB
-
IO density: IOPS/TiB: Number of input/output operations processed per second based on the total space that is being consumed by the workload, in tebibytes.
-
Service availability
-
Durability in accurate data access
-
Latency and speed
-
Capacity measurement in gibibyte (GiB), tebibyte (TiB), and pebibyte (PiB): Measurements of data storage capacity using base of 1024 (1 GiB = 10243 bytes, 1 TiB = 10244 bytes, and 1PiB = 10245 bytes).
-
Operations counter chart in IOPS/TiB: The protocol operations per second, requested by the application, divided by the size of the volume used by workloads.
-
Availability: Measured as a percentage of the number of I/O requests successfully responded to by the service, divided by total number of I/O requests made to the service. This is measured at the service demarcation in a month and does not include the scheduled service downtime or unavailability of the facilities, network, or other services provided by the customer.
-
Durability: Percentage of data accessed without loss of fidelity, excluding customer-caused deletion or corruption.
-
Latency: Time to service an I/O request received from a client, measured at the service demarcation (storage controller I/O port).
Throughput performance metrics are applicable only for file and block services based on:
-
32 KB block sizes
-
70% read/30% write I/O mix
IO density calculated in IOPS/TiB and/or MBps/TiB varies based on the following factors:
-
Workload characteristics
-
Latency, excluding the following:
-
Application latency
-
Host latency
-
Latency in the customer network while transferring data to and from the controller ports
-
Overhead latency associated with data transfer to the object store in the case of FabricPool
-
The latency automatically applied by the QoS to keep IO within service level maximums
-
-
The user and Snapshot copy data that is counted as part of the used capacity
-
The allocated absolute minimum IOPS on each ONTAP volume, regardless of the amount of data in the volume:
-
Extreme: 1,000 IOPS
-
Premium: 500 IOPS
-
Performance, Standard, and Value: 75 IOPS
-
-
While using the Advanced Data Protection add-on services, the target latency applies only to servicing IO requests from the local storage.