Data is essential in the artificial intelligence (AI) era — it is, after all, what fuels complex large language models (LLMs).

But AI is demanding and voraciously hungry, which can slow down time to production, which can be detrimental in today’s fast-paced technology landscape.

Data infrastructure leader NetApp aims to help alleviate these pressures, announcing today new AFF A-Series systems that can power complex genAI, enterprise database and cloud workloads. The new series can securely help eliminate storage silos and complexity, the company says.

“AI is a massive opportunity for companies to leverage their data in new ways to unlock competitive advantages,” said Ashish Nadkarni, group VP and GM for infrastructure systems, platforms and technologies and BuyerView Research at IDC. “However, as the AI market develops, how organizations approach AI may change. They need storage infrastructure that gives them the flexibility to combine their on-premises data storage with cloud environments.”

NetApp’s new systems include AFF A1K, AFF A90 and AFF A70. These deliver up to 2X better performance with 40 million input/output operations per second (IOP) with 1 terabyte per second throughput.

The company says this provides “proven 99.9999%” data availability. The new systems also have raw-to-effective data capacity, including always-on data reduction. NetAPP’s unified data storage supports block, file and object storage protocols and integrates with the top cloud providers. The new series also has integrated capabilities that can help optimize VMware storage costs.

The new AFF systems can provide “significant performance to feed those extraordinarily hungry workloads like AI,” Jeff Baxter, VP for NetApp product marketing, told SDxCentral.

Further, to help address today’s growing security concerns, the systems have integrated real-time ransomware detection.

“Attackers are very sophisticated,” said Baxter. A locked-down system with 99% real-time detection provides an “extra level of assurance against anyone being able to compromise that.”

The NetApp AFF A-Series systems use Intel Xeon processors to help accelerate AI adoption, said Justin Hotard, EVP and GM for Intel’s data center and AI group.

“AI is creating the biggest business transformation opportunity we’ve seen in decades, allowing enterprises to unlock new sources of value from their data,” he said.

Sandeep Singh, SVP and GM of enterprise storage at NetApp, agreed, saying: “Data is undeniably the most valuable asset for any company to outpace its competitors. Whether it's mission critical applications or leveraging enterprise data to fuel AI, the data infrastructure a company chooses to run it on makes all the difference.”

NetApp is announcing several other new features and capabilities today, including the following:

• Five new StorageGRID models to bolster the value of large unstructured data and reduce cost. The models can use capacity flash to quickly provide fast, low cost object access times and provide up to 3X performance increase, as well as 80% footprint reduction and power consumption savings up to 70%.
• Cyber vault reference architecture that extends NetApp’s data protection capabilities. This new architecture combines capabilities in secure data storage, real-time ransomware detection and provides air-gapped storage that isolates storage from unsecured networks, platforms, apps and production environments.
• SnapMirror active sync, which provides active-active continuity across two data centers. It is combined with VMware vSphere Metro Storage Cluster (vMSC) and Oracle, SAP and Microsoft databases to support ongoing, disruption-free business operations should a data center outage occur.
• FlexCache with Writeback, which creates local copies of data for distributed teams. These copies can read and write data, providing greater control and data consistency.
• BlueXP classification, which allows organizations to automatically classify, categorize and tag data across their estates. This can help enhance data intelligence as well as governance, compliance and security efforts. Classification can also support genAI and retrieval-augmented generation (RAG), which allows organizations to use their private, proprietary data.

Also today, NetApp is announcing the new NetApp AIPod with Lenovo ThinkSystem servers for Nvidia OVX. This converged infrastructure will support RAG.

Prebuilt LLMs such as ChatGPT aren’t updated with new data — in fact, they often haven’t been updated with new data since their training. RAG securely uses up-to-date, company-specific data to help organizations get the most out of their models.

Now, organizations will have the ability to conduct RAG and inferencing operations in use cases such as chatbots, object recognition and knowledge management.

NetApp AIPod combines NetApp's storage with hybrid cloud data management, including Lenovo's ThinkSystem SR675 V3 servers that support Nvidia architecture.

“As customers deploy AI, they demand business critical availability, ease of management, and infrastructure efficiency.,” said Kamran Amini, VP and GM of server, storage and software defined solutions for Lenovo Infrastructure Solutions Group.

The NetApp-Lenovo-Nvidia partnership can enable hybrid AI and make AI “more accessible for businesses of every size,” said Amini.

Bob Pette,VP of enterprise platforms at Nvidia, agreed that “making AI accessible and scalable for every industry is more important than ever.”