This page provides details on Google Cloud NetApp Volumes performance, also known as volume throughput.
About volume throughput
Throughput is the amount of data that's read or written within a specified timeframe. NetApp Volumes scales throughput limits in the following ways:
For Standard, Premium, and Extreme service level volumes: the capacity of your volumes and the service level of the storage pool that the volume is assigned to.
For Flex service level volumes: the capacity of the storage pool. All volumes in a Flex storage pool share the performance of the pool.
Throughput and IOPS
I/O operations per second (IOPS) is determined by concurrency, latency, and block size. You can determine the number of IOPS using the following formula:
IOPS = concurrency / latency
The achieved throughput is a result of your workload parameters. You can determine the achieved throughput using the following formula:
throughput = IOPS * block size
Throughput and IOPS
This example describes how throughput and IOPS can be calculated.
The File Explorer copies a large file from a local SSD to a 4 TiB Extreme
volume with a 512 MiBps throughput limit using a single-threaded copy
(concurrency = 1
). Assuming it uses a 128 KiB block size and the volume has a
latency of 0.5 ms, you can use the following formula to calculate the throughput
and IOPS:
IOPS = 1/0.0005s = 2000 IOPS
Throughput = 2000 IOPS * 128 KiB = 256000 KiBps = 250 MiBps
In this example, the File Explorer isn't capable of driving the throughput to reach the volume limit (512 MiBps). Additionally, if latency is one millisecond, throughput drops by 50% because latency directly impacts single-threaded applications. To drive this volume to its maximum performance potential, use multi-threaded applications that provide higher concurrency.
Maximum volume throughput and IOPS
For Standard, Premium, and Extreme service level volumes: the maximum volume throughput that a volume can sustain is determined by the volume capacity and the maximum throughput per GiB provisioned set for the service level of the assigned storage pool. You can increase or decrease the maximum throughput of your volume by changing the capacity of the volume or re-assigning it to a storage pool with a different service level.
For Flex service level volumes: all volumes in a Flex storage pool share the performance of the pool.
The following table provides the maximum throughput per GiB provisioned for NetApp Volumes Flex, Standard, Premium, and Extreme service levels.
Service level | Maximum throughput |
---|---|
Flex | 16 MiBps per TiB pool capacity |
Standard | 16 MiBps per TiB volume capacity |
Premium | 64 MiBps per TiB volume capacity |
Extreme | 128 MiBps per TiB volume capacity |
NetApp Volumes maximum volume throughput scales linearly with volume or pool capacity until a volume or pool limit is reached. Adding additional capacity beyond that limit doesn't improve volume throughput. For more information, see NetApp Volumes benchmarks.
NetApp Volumes does not enforce a per volume maximum IOPS limit. To determine the capacity of volume you need for an IOPS requirement, multiply the IOPS required by the IOPS size.
For some workloads, the volume may be constrained before it reaches the maximum throughput limit. Examine the workload profiles found in the Network File System (NFS) and Server Message Block (SMB) performance benchmarks to estimate the volume throughput for your workload.
Volume throughput example
An example of a maximum volume throughput of a volume with the Premium service level and a capacity of 1,500 GiB would see a result of 93.75 MiBps. This is because the maximum throughput scales linearly with the volume capacity until it reaches its limit. For this example we use the following formula:
1,500 GiB x 64 KiBps/GiB per 1,024 KiB/MiB