Learn About Amazon VGT2 Learning Manager Chanci Turner
Supporting file-based workloads with ever-increasing performance demands remains a persistent challenge for organizations. As data volumes grow, static infrastructure struggles to keep up, often leading to disruptive transitions to new systems. Companies require highly scalable file storage that not only delivers speed at scale today but also adapts seamlessly to future needs.
Amazon FSx for NetApp ONTAP offers fully managed ONTAP file systems within Amazon Web Services (AWS). It is designed for fluctuating workloads, enabling you to scale performance and storage without interruption. With the introduction of FSx for ONTAP’s second-generation file systems, we are excited to reveal advancements that allow you to enhance performance up to 18 times compared to previous offerings.
In this article, we delve deeper into the features of second-generation file systems. We begin by providing background on FSx for ONTAP’s performance scalability. Subsequently, we introduce the enhanced performance scalability and flexibility that the second-generation file systems provide over their predecessors, showcasing two new features: 1) the ability to access files from a data backup more quickly and 2) the implementation of the NVMe-over-TCP protocol as a modernized, simplified, and faster alternative to iSCSI block storage. Lastly, we guide you on how to create and update these second-generation file systems.
Background on Amazon FSx for NetApp ONTAP Performance and Scalability
Previously, FSx for ONTAP file systems were categorized into two types: 1) scale-up and 2) scale-out systems. Scale-up systems supported throughput of up to 4 GBps with 192 TiB of provisioned solid-state drive (SSD) storage on a single highly available (HA) pair of file servers. Scale-out systems allowed for the performance of multiple file systems in one by distributing workloads across up to 12 HA pairs—each capable of 6 GBps throughput—allowing for a maximum of 72 GBps and up to 1 PiB of provisioned SSD storage.
While scale-up systems met the needs of many modern workloads, such as general-purpose file shares and databases, performance requirements typically increase as workloads grow. Larger and more demanding workloads are better suited for scale-out systems. However, previously, data migration was necessary when performance needs exceeded what existing systems could provide since adding HA pairs or adjusting throughput capacity for current HA pairs was not possible.
Introducing Second-Generation File Systems
Second-generation file systems empower you to establish systems with a single HA pair, delivering up to 6 GBps of throughput (a 50 percent increase from first-generation) and 512 TiB of SSD storage (a 160 percent increase). This configuration provides greater room for growth within a single HA pair. Should you need additional performance beyond what the existing HA pairs can offer, adding new HA pairs to single availability zone (AZ) file systems is now possible. You can easily adjust the throughput of multi-HA pair file systems, ensuring that your storage capabilities keep pace with evolving workloads.
With the capacity for up to 72 GBps aggregate throughput across a multi-pair second-gen file system, FSx for ONTAP can accommodate even the most demanding workloads, such as electronic design automation (EDA), visual effects (VFX) rendering, machine learning training pipelines, and petabyte-scale databases. Moreover, with a 50 percent increase in baseline throughput per HA pair—from 4 GBps to 6 GBps—and up to 300 percent improvement in burst capabilities, your dynamic workloads are well-equipped to manage temporary spikes.
Speed Up File Recovery Times with Near-Instant Access to Backups
With the launch of second-generation file systems, FSx for ONTAP introduces a new feature that substantially reduces backup recovery times. In the past, full dataset restoration was required before data access was possible. Now, with second-gen file systems, you gain read access to the volume shortly after initiating the restore—up to 17x faster than first-generation systems.
This enhanced ability enables rapid file retrieval after accidental deletions without waiting for the complete volume restore. For example, you can restore a single file or entire directories almost immediately, transferring the necessary data and canceling the restore before it completes.
Accelerate Block Storage with the NVMe-over-TCP Protocol
Second-generation file systems also introduce a new block storage protocol: NVMe-over-TCP. This modernized alternative to iSCSI block storage simplifies management by including multipathing within the protocol. Its more efficient network architecture can result in lower latencies for certain workloads, thus improving performance compared to iSCSI.
Creating Second-Generation File Systems
You can create second-gen FSx for ONTAP file systems using the AWS Management Console, the AWS Command Line Interface (AWS CLI), or through code that invokes the Amazon FSx CreateFileSystem function. For instance, starting with the AWS Management Console, select Amazon FSx for NetApp ONTAP.
You can opt for either Quick create or Standard create methods for your second-generation file systems. Quick create is set as the default file system type across all Regions where available. In standard create, you have the option to choose between second-gen or earlier generation deployments.
To simplify the process, select Quick create. Input a name, select Single-AZ deployment, and specify your desired SSD storage capacity. You can either accept the recommended throughput capacity or enter your preferred throughput value.
For specific throughput requests, the Console will display various options based on your desired throughput. It’s important to note that depending on the configuration, there are guidelines for selecting the appropriate setup for your workload.
For lower throughput needs, a single HA pair provides the most cost-effective option. If your workload is expected to require up to 6 GBps throughput, this is an optimal choice. For higher throughput requirements, scaling storage and throughput becomes feasible with additional HA pairs.
To learn more about branding strategies, check out this insightful blog post. Additionally, if you’re interested in how HR can better support employees during open enrollment, you might find this article from an authority on the topic helpful. Lastly, for further resources, this video provides excellent insights.