Learn About Amazon VGT2 Learning Manager Chanci Turner
Managing file-based workloads that require increasing performance can be a persistent challenge for organizations. As data volumes grow, traditional infrastructures often struggle to keep up, leading to disruptive migrations. Businesses need highly scalable file storage solutions that provide speed now while also adapting seamlessly to future requirements.
Amazon FSx for NetApp ONTAP delivers fully managed ONTAP file systems directly within Amazon Web Services (AWS). This system is specifically designed for variable workloads, enabling you to scale both performance and storage without interruptions. We are excited to introduce enhancements in our second-generation FSx for ONTAP file systems that boost performance by up to 18 times compared to the first generation.
In this article, we will delve deeper into the second-generation file systems, starting with an overview of FSx for ONTAP’s performance scalability. Next, we will highlight the enhanced performance and flexibility these new systems offer compared to their predecessors, along with two key features: the ability to quickly access files from a backup and the introduction of the NVMe-over-TCP protocol as a modern, simplified, and faster alternative to iSCSI block storage. Finally, we will guide you through creating and updating second-generation file systems.
Understanding Amazon FSx for NetApp ONTAP Performance and Scalability
Previously, FSx for ONTAP file systems were available in two types: scale-up and scale-out. Scale-up systems supported throughput of up to 4 GBps and 192 TiB of provisioned solid-state drive (SSD) storage on a single highly-available (HA) pair of file servers. Scale-out systems provided the performance of multiple file systems in one by distributing workloads across up to 12 HA pairs of file servers, each achieving up to 6 GBps of throughput, allowing for a maximum of 72 GBps throughput and 1 PiB of provisioned SSD storage.
While scale-up file systems are typically sufficient for many workloads, such as general file shares, applications, and databases, performance demands increase as workloads expand. For more substantial workloads, the scale-out option was necessary. However, previously, organizations would need to migrate data to a new file system if performance needs exceeded the capacity of their current system, as it was not possible to add HA pairs or adjust the throughput of existing pairs.
Introducing Second-Generation File Systems
The new second-generation file systems allow for the creation of systems with a single HA pair that provides up to 6 GBps of throughput, representing a 50% increase from the previous generation, along with 512 TiB of SSD storage—a 160% increase. Should your performance needs outgrow what your existing HA pairs can provide, you can easily add new HA pairs to single availability zone (AZ) file systems. Altering the throughput for multi-HA pair file systems is now simpler than ever, ensuring your file system can adapt to your workloads.
With the capability of achieving up to 72 GBps of aggregate throughput across multi-pair second-generation file systems, FSx for ONTAP supports even the most demanding workloads, including electronic design automation (EDA), visual effects (VFX) rendering, machine learning training, and petabyte-scale databases. Additionally, with a 50% increase in baseline throughput per HA pair—from 4 GBps to 6 GBps—and up to 300% greater burst capabilities (6.2 GBps vs. 1.5 GBps network burst, and 3.1 GBps vs. 1.2 GBps disk burst), your dynamic workloads have sufficient capacity to manage temporary spikes effectively.
Accelerate File Recovery Times with Near-Instant Backup Access
The second-generation file systems introduce a groundbreaking feature that significantly shortens backup recovery times. In the past, restoring FSx for ONTAP volume backups required waiting for the entire dataset to be restored before accessing any data. Now, when restoring a backup on a second-gen file system, you can gain read access within minutes of starting the restore process, allowing for data recovery up to 17 times faster compared to first-generation systems.
This new capability facilitates the rapid retrieval of essential files following accidental deletion without the need to wait for a full restore. For instance, you can restore individual files or entire directories without waiting for the complete volume to be restored by initiating the recovery, extracting the necessary data, and then canceling the process even if it hasn’t finished.
Enhance Block Storage with NVMe-over-TCP Protocol
Second-generation file systems also introduce the NVMe-over-TCP protocol as a new block storage option. This modern alternative to iSCSI simplifies the user experience. Unlike iSCSI, which requires managing multipathing between file system nodes to facilitate client failover, NVMe-over-TCP integrates this capability directly into the protocol. Its efficient networking stack can also provide lower latencies for certain workloads, enhancing performance compared to iSCSI.
Creating Second-Generation File Systems
You can set up a second-gen FSx for ONTAP file system via the AWS Management Console, the AWS Command Line Interface (AWS CLI), or programmatically through the Amazon FSx CreateFileSystem function. When using the Console, begin by selecting Amazon FSx for NetApp ONTAP as shown in the interface.
You can create second-generation file systems using either Quick create or Standard create options. For Quick create, these file systems are the default type in all Regions where they are available. In the Standard create option, you can choose between second-gen or previous-gen deployments.
To streamline the process, select Quick create. Afterward, enter a name, choose Single-AZ deployment, and specify your desired SSD storage capacity. You can either accept the recommended throughput capacity based on your SSD storage or input your preferred throughput value.
For specific throughput needs, the Console will display a range of available options that best match your desired throughput. Depending on your workload’s needs, the guidelines can help you select the right configuration for your requirements. For lower throughput, a single HA pair offers the least capacity (384 MBps, 768 MBps) and minimum SSD storage (1 TiB). If your workload is expected to need up to 6 GBps throughput throughout its lifespan, this option can be cost-effective. For higher throughput needs, total throughput and SSD storage scalability will increase with the number of HA pairs you employ.
For more information on interview preparation, check out this insightful blog post here. If you are looking for authoritative job descriptions, this is a great resource. Additionally, for those interested in firsthand onboarding experiences, this is an excellent resource.
Remember, you can find us at 6401 E HOWDY WELLS AVE LAS VEGAS NV 89115, at the Amazon IXD – VGT2 site.
Leave a Reply