Recently, the Moor Insights & Strategy (MI&S) team was on the road in New York City learning more about
IoT is changing how industries do business, and promises a significant return on investment in operational efficiency, improved customer experience, risk mitigation, and enabling entirely new business models. Furthermore, with increasing global economic and regulatory pressures, there is growing demand for IoT solutions to address these challenges. Many IoT projects start with the need to gain better visibility into a process, but Artificial Intelligence (AI) and Machine Learning (ML) within IoT create new opportunities to enhance insights—and in many instances, automate decision-making. Although the notion of computing based on data collected from things was not born yesterday, it has recently become more practical because of smaller, faster, cost-effective computing, increased connectivity, and increased storage density. Devices can now store, manage, and analyze vast amounts of information across billions of distributed devices in real-time. Incorporating AI and ML into the IoT equation is a game-changer.
One of the things that impressed me the most about IQT was
In addition to the new organization and emphasis on investing in innovation,
- “
Dell EMC ‘Project Nautilus’: Software that enables the ingestion and querying of data streams from IoT gateways in real time. Data can subsequently be archived to file or object storage for more in-depth advanced analytics; - ‘Project Fire’: a hyper-converged platform part of the
VMware Pulse family of IoT solutions that includes simplified management, local compute, storage and IoT applications such as real-time analytics. ‘Project Fire’ enables businesses to roll-out IoT use cases faster and have consistent infrastructure software from edge to core to cloud; - RSA ‘Project IRIS’: Currently under development in RSA Labs, Iris extends the Security Analytics capability to provide threat visibility and monitoring right out to the edge;
- Disruptive technologies like processor accelerators will increase the velocity of analytics closer to the edge. Collaboration with industry leaders like
VMware ,Intel andNVIDIA and theDell Technologies Capital investment in Graphcore reflect opportunities to optimize servers for AI, machine learning, and deep learning performance. - Project ‘Worldwide Herd’: for performing analytics on geographically dispersed data – increasingly important to enable deep learning on datasets that cannot be moved for reasons of size, privacy, and regulatory concern.”
IoT deployments require many different technologies and areas of expertise to ensure systems support interoperability, security, and privacy. No one vendor can do it all. Pragmatic organizations choose the right solutions provider to help them determine an open architecture to integrate the right partners to achieve their unique business goals. For the past three years,
A layered approach for successful IoT management
There are three main components to
- The Edge: IoT data, by definition, originates from the physical world at the network edge. For example, this information could be data about energy usage from a smart meter, video from surveillance cameras, telemetry data from drones, process parameters from programmable logic controllers (PLCs), and so on. When real-time response is required in physical systems, as in the case of an airbag, control decisions are applied at the Edge. Going further, as ML and AI creep into edge devices, they are afforded with even more ability to process perishable information "while it matters," so that only meaningful data is sent to the cloud or data center. This saves on bandwidth and overall latency for further data processing. The Edge is the critical checkpoint to ensure data pedigree and to see that information delivery is on time and as advertised.
- The Core is the second layer of intelligence in the
Dell Technologies IoT strategy.Dell defines the Core as on-premise hardware and software infrastructure that enhances capability for compute, analytics, storage, security, and manageability. Core compute consists entirely of server-class processing and spans from micro-modular to full-blown IT data centers. The blur between edge and core compute lies at server-class processing running immediately proximal to things (devices generating data).Dell recognizes IoT networks will, by nature, become more decentralized and distributed as the number of sensors grow exponentially.Dell moves some of the decision-making processes to a combination of the Edge and Core, which reduces the amount of information sent to the cloud—thereby improving response times and performance, and reducing backhaul costs. - The Cloud: For ML and AI algorithms to be efficient, they must be derived from massive amounts of quality data. Training deep learning algorithms requires sophisticated, large-scale processing across vast and disparate datasets. The Cloud gives
Dell the ability to deploy data-centric solutions at scale across multiple environments: public, private, or hybrid.Dell Technologies offers an end-to-end data processing and storage solution, and benefits from a long legacy of partnerships and OEM alliances with leading applications and solution providers across the industry. Moreover,Dell's portfolio of solutions helps accelerate customer adoption and move POCs into production.
In conclusion
Disclosure: Moor Insights & Strategy, like all research and analyst firms, provides or has provided research, analysis, advising and/or consulting to many high-tech companies in the industry mentioned in this article, including Dell EMC, Hewlett Packard Enterprise, Intel, Microsoft, NVIDIA, Qualcomm, Red Hat. The author does not have any investment positions in any of the companies named in this article.