WTL

How to realise the lowest latencies for your enterprise applications

latency

When it comes to artificial intelligence and machine learning applications, every microsecond counts. As well as ensuring you have sufficient processing power, it is essential that you remove latency from every layer of your technology stack.

As you would expect when dealing with enormous data sets, choice of storage is critical. For cutting edge AI models to operate optimally, you will need cutting edge storage to underpin it. And this is where the NetApp AFF A-series comes into its own.

All Flash Arrays – obviously

Flash storage currently outperforms any other medium, so it is the natural choice for mission-critical operations. As the name implies – AFF – the NetApp A-series is an all-flash array, using high speed SSD drives to deliver up to 1.4 million IOPS.

No other storage medium is suitable for processing mission-critical applications involving large data sets – up to 702.7PB with the NetApp AFF family. And with native de-duplication, compression and compacting, these systems ensure your storage space is maximised, increasing return on investment and lowering total cost of ownership.

NVMe for maximum throughput

You can further reduce latency by accelerating data transfer between storage and CPU. The NetApp A-Series systems offers NVMe/TCP for connection to existing ethernet infrastructure. Alternatively, double IOPS potential and halve latency the latency found in traditional channel connections with NVMe/FC. Whichever you choose, NVMe connectivity will also allow you to attach ultra-high-performance storage to your existing SAN without disrupting operations to achieve up to 300GB/s throughput.

Flexible connectivity options

In the modern hybrid cloud operating environment, your business needs seamless connectivity to on- and off-premises systems. In addition to high-speed NVMe, the A-series also offers a complete range of alternative technologies including iSCSI, FCoE, NFS, SMB, Amazon S3.

This means that A-series systems will slot directly into your existing infrastructure. You can take advantage of ultra-high performance SSD storage for local processing and off-load lower priority workloads to low-cost secondary systems or cloud platforms.

Scalability

In addition to support for external services, the NetApp A-series is also fully scalable. As your demands increase, the system scales to 24 nodes for a maximum total capacity of 702.7PB of high resilience, high performance all flash storage.

And as you would expect, the A-series is fully supported by ONTAP software to simplify the process of managing and optimising your storage workloads. ONTAP allows you to scale at speed whenever your applications demand.

Find out more

Ready to learn more about the NetApp AFF A-series and how it will help your business achieve its high-performance computing objectives? Get in touch today and our team of storage experts will guide you through the benefits and features so you understand where the A-series fits into your storage strategy.