Competitive Advantage through Artificial Intelligence with NetApp Data Fabric
Under the scientific direction of Cambridge Consultants, a new product was created from Nvidia and NetApp components: NetApp Ontap AI. This hardware is the world's first AI-enabled accelerator for data management and transport.
The reasoning behind this: The faster neural networks can be supplied with data for machine/deep learning, the faster and better these AI algorithms will work.
By its very nature, a deep learning algorithm for optimal pricing only makes sense if the results are available in real time, i.e., at the time when the price point is set. A system that needs to determine the optimal price for a seasonal item should be through with the learning algorithm before the end of the season.
The bottleneck for many neural networks in the learning and testing phase is data provision. In an e-commerce system, sales data can be available on- and off-premise. A "data lake" consolidation and a fast transport to the deep learning system are basic requirements for sustainable success.
In a collaboration between AI specialist Nvidia and data storage specialist NetApp, a hardware rack containing components from Nvidia and NetApp was created with scientific advice from Cambridge Consulting.
This hardware combination handles data transport management for the learning and test phase of neural networks, among other things. Christian Lorentz, senior product and solution marketing manager at NetApp EMEA, presented the Nvidia/NetApp hardware at Cambridge Consultants (UK) in September of this year. Development is in its early stages and all parties are currently evaluating further applications and future expansion stages.
This should enable the creation of a hybrid cloud architecture for IoT. NetApp provides data management capabilities and solution support to address the ever-growing data sources with virtually unlimited scalability and performance.
In this way, data-intensive applications in the field of AI/AI, machine learning and deep learning can be supplied with data, trained and executed. In the context of an SAP ERP and SAP Leonardo installation, there are already numerous possible applications.
From the Leonardo IoT area, there is the current requirement for more data throughput and low latency. Terabytes of data are generated at the edges of the network by the millions of IoT sensors. At the moment, attempts are being made to tame this flood of data with edge computing.
Future AI algorithms for data analysis, compression, and routing based on NetApp Ontap AI could provide significant relief from a critical situation today.
At the other end of an end-to-end data scenario, we find multi-cloud and hybrid cloud. In an on-premise and off-premise ERP system, fast, efficient, resource-efficient data traffic takes on a very special significance because sufficient bandwidth is not always available. Here, NetApp Ontap AI could not only reduce the data load, but also optimize the data traffic with AI algorithms.
NetApp helps organizations more broadly target artificial intelligence (AI/AI) through easily deployable data services, from the point of origin to the data center to the cloud.
NetApp is one of the leading specialists for data management in the hybrid cloud and also presented five new solutions and services: Ontap 9.5, Max-Data, StorageGrid SG6060 and the Flash Performance Guarantee.
These new data services and solutions expand the scope of NetApp's data fabric concept for managing data across the board between its point of origin, the central data platform, and the cloud.
This enables companies to fully exploit the potential of artificial intelligence. According to Gartner, global AI business value will total $1.2 billion this year, and this represents a 70 percent increase over 2017.
IDC predicts that nearly 75 percent of commercial enterprise applications will use AI/AI by 2021. Many companies are looking to leverage artificial intelligence for new and innovative business opportunities related to diverse, distributed and dynamic data sets.
However, they often find that the most valuable data is locked away in silos. This usually makes access too complex and costly for resource-intensive AI/AI applications like SAP Leonardo.
"Today, enormous amounts of data from IoT devices and sensors are generated and managed in on-premises data centers and in hybrid cloud environments. Therefore, a data fabric architecture that covers the point of origin, the central data platform and the cloud is essential for the successful deployment of AI"
said Joel Reich, EVP, Storage Systems and Software at NetApp.
"When companies integrate the new data services and solutions using the NetApp Data Fabric concept, they can accelerate data processes for the entire enterprise."
Storage Optimization for AI Data Management
NetApp Ontap 9.5 software helps enterprises modernize their data services. The software offers leading cloud integration, optimal all-flash performance, and increased efficiency and ease of use.
Ontap 9.5 offers versatility with high performance as well as consistently low latency. Critical workloads can be accelerated with end-to-end NVMe capabilities.
Furthermore, Ontap 9.5 automates the tiering of petabyte-scale data sets to object stores in the cloud or on-premises and reduces implementation costs for multi-site business continuity solutions.
Max-Data enables customers to accelerate the performance of their applications without changing code. Max-Data is the industry's first solution to use persistent memory in servers to deliver memory-like low latency and Flash-like capacity.
Accordingly, Max Data achieves higher performance for application-level data, enabling faster processing of data for AI/AI applications and real-time data analytics.
Furthermore, Max-Data accelerates application data recovery with data protection services for persistent storage on servers and increases efficiency by requiring fewer servers for equivalent or even better performance.
NetApp's Flash Performance Guarantee provides the industry's first latency guarantee, enabling customers to run AI/AI applications at a predictable low latency.
This performance guarantee is unique in the industry and offers a constant latency of no more than 0.5 milliseconds. This enables companies to minimize costs and risks by using flash.
NetApp StorageGRID SG6060 object storage with flash acceleration meets the needs of performance-intensive, concurrent workloads for IoT and Big Data analytics environments and machine learning systems.
Combined with the Ontap fabric pool capabilities, companies have excellent data management capabilities. At the same time, they benefit from the cost-effectiveness of object storage.