Innovation Monitor: Gartner Hype Cycle Trend #5 — Composite Architecture

Innovation Monitor: Gartner Hype Cycle Trend #5 — Composite Architecture

View this email in your browser

Welcome to this week’s Innovation Monitor. We’re focusing on a business trend: composite architecture, that Gartner defines in terms of the “composable enterprise” that’s designed “to respond to rapidly changing business needs with packaged business capabilities built upon a flexible data fabric…. [and] built-in intelligence is decentralized and extends outward to edge devices and the end user.”

Imagine a business that can shift strategy and direction in a near-instantaneous manner, responding to on-the-ground data. You have a five year plan, but you also have a five minute plan. Composite architecture represents building organizations capable of delivering on this promise; that collect real-time data at every corner of their operations which feed back into smart, adaptive systems that create real-time change.

In a February whitepaper, Gartner expanded on the idea of packaged business capabilities, or PBCs. A composite architecture allows the assembly, reassembly, and extension of PBCs. (For example, a shopping cart in an ecommerce app or an expense approval mechanism in a business line could be a PBC). These building blocks can be either sourced from third parties (software vendors), developed internally, built from legacy systems (or as Gartner elegantly puts it, “the modernization of monolithic applications into more modular components”), or created/assembled by technically skilled employees (something that’s becoming more prevalent with enterprise codeless).

Think of it like a constantly shifting and evolving organization, built on an infrastructure that’s malleable and flexible. Here’s perhaps a pretty perfect GIF that captures this idea :)

Composite architecture, in a nutshell, promises to help businesses respond to rapidly changing business needs (like during a pandemic or recession). Let’s break the concept into three distinct technology categories we can explore:
1) data fabric,
2) embedded AI, and
3) single-board, 5G computers at the edge.

As always, we wish you and your community safety, calm and solidarity as we support each other through this unprecedented time. Thank you for reading!

All best,
Erica Matsumoto DATA FABRIC What Is a Data Fabric?

Donald Feinberg, VP at Gartner, said that “the size, complexity, and distributed nature of data, speed of action and the continuous intelligence required by digital business means that rigid and centralized architectures and tools break down.” A data fabric is meant to enable “frictionless access and sharing of data in a distributed environment” — aka the foundation that supports the PBC model. Data in a data fabric is meant to be “easy to explore, analyze and understand,” and can provide a “catalog of consistent data services across private and public clouds.”

In an enterprise, data might be stored in relational databases, flat files, data lakes, data stores, and more. There is also a multitude of applications, platforms, and data types to deal with. Managing access and security across organizational data becomes difficult and cumbersome. A data fabric “consolidates data management into one environment, automatically managing disparate data sources and technologies in both on-premises and cloud environments.” One example is IBM Cloud Pak for Data.

AtScale — 5 min read Read More How Do You Weave a Data Fabric?

To better understand how a data fabric is actually “weaved”, The New Stack spoke to Anthony Lye, senior vice president and general manager of the cloud business unit at NetApp, Jack Norris, Senior Vice President, Data and Applications at MapR Technologies, and Isabelle Nuage, director of product marketing at Talend.

Nuage noted a few key characteristics of a data fabric:

• “It connects to anything via pre-packaged connectors and components.”
• “Offers built-in machine learning, data quality, and governance capabilities.”
• “Enables data integration and application integration scenarios.”

Accessible data, in turn, enables capabilities across big data analytics, predictive maintenance, risk, and fraud prevention. It also lets organizations do two seemingly contradictory things: open up data access to the public while keeping things secure.

Real-world use case: Domino’s Pizza now allows you to order from a smartwatch, smart TV, your social account, or your car. “This created 17TB of data, from 85,000 data sources, both structured and unstructured,” so the company used a data fabric to weave data from their POS systems, supply chain centers, and all digital channels.

But what does the structure look like? Norris wrote Twelve Rules for Data Fabrics, and that’s worth a deeper look if you’re interested. Lye breaks it down into four simple layers:

• “The lowest layer is data storage. Here, there’s a set of APIs that enable the management of protocols up to customers through user interfaces.”
• “Next is data services. APIs help manage several services at this level, including data protection, moving data, securing the data and inspecting and classifying the data.”
• “The next layer is the control plain, which includes tools to circle the data up to the people who will use them, typically the Site Reliability Engineer (SRE).”
• “The top layer is analytics.”

The New Stack — 7 min read Read More EMBEDDED AI The Future of Embedded AI and Custom SoCs

Compute is moving to the edge — and that includes expensive operations like AI inference. Systems on a Chip (SoCs) are starting to embed more advanced AI capabilities, allowing for processing to happen on-device rather than in the cloud. The linked NWES blog goes into the technical details, but the main takeaway is the industry is moving towards fully-customized SoCs capable of powerful AI tasks.

The reasons are numerous. For Apple, which spent $200M acquiring low-power ML software and hardware startup Xnor.ai, edge AI means fast on-device photo processing, community apps that can take advantage of AI, and better security (since data is processed on-device rather than being sent to the cloud). Edge devices can operate in remote locations without strong network access and consume less power — think medical devices in remote regions. In a retail business, AI-enabled sensors and cameras might track shopper activity and feed the data back to the cloud for further post-processing and analysis.

NWES — 5 min read Read More AI Is Making Robots More Fun

The article is primarily about — surprise — robots. But this paragraph in particular is relevant: “Advanced entertainment robots are much more than toys. Embedded AI technologies such as computer vision and natural language processing enable the robots to identify and track human faces, receive and recognize voice commands. They can respond and interact with behaviour like that of a house pet, or, when integrated with IoT modules and programmable systems, function as a humanoid Alexa personal assistant. Such robots are also intriguing and engaging students in educational roles. Across applications, an overarching trend is to make these front-line machines more fun for humans to deal with.”

Synced — 4 min read Read More SINGLE BOARD COMPUTERS + PRIVATE 5G What Is Private LTE and Private 5G

Before we dive into 5G’s role in Single Board Computers (SBCs), let’s define the difference between SoCs and SBCs: the former can be a component of the latter — SBCs are fully functioning computers (think Rasberry Pi or Nvidia’s Jetson Nano). SoCs are more customizable (or may be too narrow), while SBCs are easier to use (or may be too generic). Embedded AI and SBCs go hand-in-hand.

Gartner analyst Tony Harvey says that “IT managers should look for single-board edge servers that can be rolled out rapidly and be managed and updated in the field without skilled staff on-site. Security should be built into the system across all areas including, physical, data storage, communications, management, and updates.”

Ok, let’s wrap things up with 5G and the linked article. The important thing to note here is that “5G was designed specifically to support the types of massive senor grids and industrial IoT (IIOT) applications we are currently witnessing the emergence of. A 5G radio access network (RAN) not only survives in harsh industrial environments, it thrives in them.” And the paragraph that ties the above concepts together:

“Private 5G can deliver ultra low latency and incredibly high bandwidth connections supporting artificial intelligence-driven applications serving an exploding number of sensors and endpoints. With distributed cloud architectures, local processing allows machine learning algorithms to be applied on massive amounts of data without leaving the security and privacy of the enterprise campus. Remaining in the private domain, less demanding real-time and non-real-time traffic can be offloaded to edge or core computing power, hosted by managed service providers.”

Metaswitch — 3 min read Read More This Week in Business History

This Week in Business History

September 30, 1452: The first section of the Gutenberg Bible was finished in Mainz, Germany.

It is unclear when Johannes Gutenberg conceived of his Bible project, though he was clearly in production by 1452. He probably produced about 180 copies — 145 that were printed on handmade paper imported from Italy and the remainder on more luxurious and expensive vellum. Only four dozen Gutenberg Bibles remain, and of these only 21 are complete.

This email was sent to <<Email Address>>
why did I get this? unsubscribe from this list update subscription preferences
NYC Media Lab · 370 Jay Street, 3rd floor · Brooklyn, New York 11201 · USA