edge Archives - Business Intelligence Info

Tag Archives: edge

Predictive Analytics in Manufacturing: A Winning Edge

February 1, 2020 Sisense

The modern manufacturing world is a delicate dance, filled with interconnected pieces that all need to work perfectly in order to produce the goods that keep the world running. In Moving Parts, we explore the unique data and analytics challenges manufacturing companies face every day.

Building an accurate predictive analytics model isn’t easy. It requires a skilled data team, advanced tools, and enormous amounts of clean data from the right combination of inputs. It’s a difficult process, but an effective predictive analytics engine is an enormous asset for any organization.

Manufacturing Data banner 770X250 Predictive Analytics in Manufacturing: A Winning Edge

Big challenges, big rewards

Manufacturing companies are in a unique position regarding data: they create and capture tons of it every day. The process of producing goods is an enormous opportunity for data optimization. Raw materials need to be ordered, received, constructed, packaged, and shipped out for sale in the most efficient manner possible. Because the steps are repeated so many times through the process, a small edge created via predictive analytics in manufacturing will be magnified at every repetition to produce significant benefit.

Because of the cyclical nature of the manufacturing process, data-driven companies are building superior processes to create bigger and bigger advantages. Here are a few examples of companies using manufacturing analytics to win the future:

Predicting return rate

Skullcandy’s dive into predictive analytics started with the challenge of understanding return rates on new products. The logic was that if the team could predict certain features or aspects of a product that would lead to a return, they could optimize those policies around returning products. They used BigSquid to blend and analyze historical data related to returns and added their learnings to features and products on their roadmap. From there, the team could ask new questions of that dataset to understand the way customers were interacting with their products and ultimately build a better warranty policy for products before they were even released. This data was also useful for product managers, giving them a clear picture of what was making customers adopt Skullcandy’s products (or not).

Once the return rate questions were answered, the team focused their efforts on uncovering insights around reviews and warranty claims to generate insights about positive and negative drivers. Data like this is ideal for making decisions for product roadmaps. All those customer insights can be used in a number of creative ways to better focus resources and improve products.

Improve forecasts and maximize revenue

Gentex made the most of their budget to optimize their incoming revenue. Just six months after implementing predictive analytics, their ecommerce sales increased by 50%!

Gentex deployed Sisense to comb through millions of records after switching over from an outdated ERP system. They needed a platform that could churn through all that data quickly and deliver quick intelligence about both current and future revenues. Initially, Gentex created dashboards for their sales and operations teams that collected information about sales, quotes, and orders across the company. Those dashboards answered immediate questions about the current state of the business.

To answer more forward-looking questions, Gentex creates a sales forecast for an entire year using just a few months of data. They use predictive models to forecast revenues based on spending. They even incorporate trend data to improve accuracy over time. Currently, Gentex builds visualizations of year-to-date revenue data to forecast up to 15 months into the future.

Operating off those accurate forecasts, Gentex made the most of their budget to optimize their incoming revenue. Just six months after implementing predictive analytics, their ecommerce sales increased by 50%!

Improve inventory management with demand forecasting

Making a product that consumers want to buy is only useful if a company can find a way to get that product in front of the consumers who demand it. Several of today’s most cutting-edge manufacturers are blending historical customer data and external factors to predict demand for goods so they can increase production when demand will be high and decrease production when demand will be low. These companies aren’t just building for the future, they’re building the future.

The need to accurately forecast demand is crucial to these manufacturers. Assessing demand in real-time is ineffective since companies need to make decisions about demand far enough in advance to complete an entire production cycle and get that product in front of customers. With a solid predictive analytics model in place, manufacturers can create exactly the right amount of products (and the right variety of those products) to satisfy future customers.

These forecasts optimize sales revenue, but it also avoids unnecessary costs associated with producing, shipping, and stocking items that won’t sell. Accurate predictions are a win-win for any manufacturer.

Build your manufacturing business with analytics

Predictive analytics in manufacturing have gone from being science fiction to being a make-or-break addition to any company’s technology stack. Using a platform like Sisense for manufacturing analytics, combining internal and external information into a series of accurate forecasts is incredibly invaluable to any manufacturer. Improving any step of the manufacturing process is an advantage over the competition, but improving every step is a data-driven way to become an industry leader faster.

Predictive Analytics in Manufacturing: A Winning Edge

Adam Bonefeste is a veteran content marketing manager. When he isn’t writing copy, he’s probably reading books, running through San Francisco or getting lost in YouTube holes about math/logic problems.

Let’s block ads! (Why?)

Blog – Sisense

Taking A Peek Over The Mountain Edge

January 5, 2020 Humor

A stunning waterfall.

“Over the Edge.”
Image courtesy of https://imgur.com/gallery/mOfTxBE.

Let’s block ads! (Why?)

Quipster

Get 2020 vision about edge computing and 5G

December 20, 2019 Big Data

This article is part of the Technology Insight series, made possible with funding from Intel.
________________________________________________________________________________________

With 2020 predictions looming, there’s sure to be a fresh wave of hype around the edge and 5G. Now’s an ideal time to solidify and update your understanding, and explore how they’ll complement each other. If you’re processing payments, taking online orders, detecting fraud, in the financial services industry, or exploring machine learning, these two technologies can help keep you competitive in the coming months.

What and why

Edge computing is all about processing information from devices closer to where it’s being created, rather than shuttling it back and forth from the cloud. Together with 5G, computing at the edge paves the way for applications that wouldn’t have been possible before. Think augmented and virtual reality, where ultra-low latency keeps what you see in sync with what you do, or autonomous vehicles that need to make split-second decisions based on huge volumes of data.

Globally, IDC forecasts 150 billion connected devices (including RFID) by 2025, many of which will pump out data in real time. In 2017, real-time data was a mere 15% of all information created, captured, or replicated. In 2025, it’s expected to reach 30%. As a percentage, that might not sound transformational. But it’s an order of magnitude higher in raw capacity (from ~5 to ~50 zettabytes). Edge computing gives you the power to perform intelligent analysis based on that deluge of real-time data almost instantaneously, all while minimizing your bandwidth expenses.

Definition emerging

Yes, it’s nearly 2020. But the definition of “edge” still often varies depending on who you ask. From NIST to IEEE, models are still evolving. It could be the Raspberry Pi in a refrigerated big rig selectively sending sensor information to the cloud or the node processing game data for Google’s streaming Stadia platform. Although miles of proximity separate those two interpretations, they both put compute resources closer to the user.

Thanks to the collaborative, vendor-neutral 2018 State of the Edge report, there now exists a more explicit definition with some industry consensus behind it:

The edge is a location, not a thing.
There are lots of edges, but the edge we care about is the edge of the last-mile network.
This edge has two sides: an infrastructure edge and a device edge.
Compute will exist on both sides, working in coordination with the centralized cloud.

For clarity, the device edge includes end-points like phones, drones, AR headsets, IoT sensors, and connected cars; gateway devices like switches and routers; and on-premise servers. They’re all on the downstream side of your last-mile cellular or cable network. The infrastructure edge exists on the upstream side. That’s where you find your compute resources collocated with network access equipment and regional datacenters.

Get 2020 vision about edge computing and 5G

Above: Edge computing exists along a spectrum, from the device edge to the infrastructure edge.

Image Credit: Intel

In our big rig example, the Raspberry Pi is on the device edge. Rather than chewing up bandwidth by continually transmitting environmental data, it processes locally and only phones home in the event of an emergency. Conversely, hyper-local datacenters streaming a 4K gaming experience at 60 frames per second live on the infrastructure edge. Although the device edge offers a narrow latency advantage, you’re obviously going to find much more powerful hardware further upstream.

Beyond the distributed infrastructure edge lies the core network and, ultimately, the cloud, which is more centralized and scalable. But by the time you get to the cloud, latency is much higher (and far less consistent).

Benefits of low-latency edge computing

It’s easy to point at low latency as the killer application of edge computing, particularly as cloud-based software strains under the limitations of physics. Data cannot move any faster than the speed of light, so requests to servers hundreds or thousands of miles away inevitably take tens or hundreds of milliseconds to fulfill. The difference isn’t perceivable as you scroll through your Twitter feed. But those numbers wouldn’t be acceptable to a surgeon operating remotely or a gamer in virtual reality. Above all else, processing at the edge shaves away latency to keep data relevant.

Above: Today’s Internet is ill-equipped for tomorrow’s applications, which will be bandwidth-hungry and latency-sensitive.

Image Credit: State of the Edge 2020

Edge computing also saves you from shuttling every bit of data back and forth between connected devices and the cloud. If you can determine the value of information close to where it’s created, you can optimize the way it flows. Limiting traffic to just the data that belongs on the cloud cuts down on bandwidth and storage costs, even for applications that aren’t sensitive to latency.

Reliability stands to benefit from edge computing, too. A lot can go wrong between the device edge and centralized cloud. But in rugged environments like offshore platforms, refineries, or solar farms, the device and infrastructure edges can operate semi-autonomously when a connection to the cloud isn’t available.

Distributed architectures can even be a boon to security. Moving less information to the cloud means there’s less information to intercept. And analyzing data at the edge distributes risk geographically. The endpoints themselves aren’t always easy to protect, so firewalling them at the edge helps limit the scope of an attack. Further, keeping data local may be useful for compliance reasons. An edge infrastructure gives you the flexibility to limit access based on geography or copyright limitations.

The edge is pervasive; 5G makes it better

Edge computing is not new. As far back as 2000, content delivery networks were being referred to as edge networks. But it’s universally accepted that as 5G coverage grows, edge computing is going to help address the high-bandwidth, low-latency requirements of modern applications with local, rather than regional, compute resources.

The technology underlying 5G will add speed, reliability, and flexibility to enterprise applications by getting compute resources closer to where data is being created. Information will move between 5G networks efficiently, rather than requiring a trip to the centralized cloud and back. As a result, we’re going to be looking at use cases that previously weren’t possible.

According to the 2020 State of the Edge report, the largest demand for edge computing comes from communication network operators virtualizing their infrastructure and upgrading their networks for 5G. Mobile consumer services running on those networks are going to rely on edge computing to enable streaming game platforms, augmented/virtual reality, and AI.

Smart homes, smart grids, and smart cities all share a proclivity for device edge platforms. As those use cases evolve and become more sophisticated, though, there will be demand for infrastructure edge capabilities, too. 5G’s provisions for ultra-reliable low-latency communications (URLLC) and massive machine-type communication (mMTC) mean the devices and edge can be even closer together, making their short connection more efficient.

Above: Traffic lights are networked and connected to edge gateway server, where their data can be collected and analyzed. As part of an edge network, they can feed data to map tools and reroute around congestion.

Image Credit: Moor Insights & Strategy

And who could forget about the autonomous automobile, the poster child for edge computing enhanced by 5G? Modern automobiles already utilize compute resources on the device edge for collision avoidance, lane-keeping, and adaptive cruise control. But as assisted and autonomous driving features become more sophisticated, infrastructure edge resources will be required to add intelligence that could only come from the surrounding environment. Good examples: rerouting a trip based on traffic miles ahead, communicating with other autonomous vehicles to accelerate from a stoplight in unison, or making split-second decisions to avoid unsafe situations.

The edge is still young

The empowered edge is one of Gartner’s Top 10 Strategic Technology Trends for 2020. However, several other concepts on its watchlist have roots in edge computing as well. Hyperautomation, which deals with the application of artificial intelligence and machine learning to augment human input, is going to rely on a foundation of low latency and unflagging reliability. The multiexperience is another example, incorporating multisensory and multitouchpoint interfaces dependent on high bandwidth and real-time processing. Of course, autonomous things are all about AI, 5G, and edge computing.

Above: The global annual CAPEX for edge IT and data center facilities is forecast to reach $ 146 billion in 2028 with a 35 percent CAGR.

Image Credit: State of the Edge 2020

Enabling those novel use cases will require substantial investment. A forecast model by Tolaga Research predicts a cumulative CapEx spend of $ 700 billion between now and 2028 on edge IT and datacenter infrastructure. As the computing pendulum swings from the centralized cloud to a distributed edge, opportunities abound, particularly in the maturing infrastructure edge. Understanding the impact of edge computing and 5G will allow you to provide seamless customer experiences, test new markets, and act on insights in real-time.

Let’s block ads! (Why?)

Big Data – VentureBeat

Google’s Coral AI edge hardware launches out of beta

October 23, 2019 Big Data

Last March, Google took the wraps off of Coral, a collection of hardware development kits and accessories intended to bolster the development of machine learning models at the edge. It launched in select regions in beta, but the tech giant today announced that it’s graduating to a “wider” and global release.

All Coral products — including the $ 150 Coral Dev Board, the $ 74.99 Coral USB Accelerator, and the $ 24.99 5-megapixel camera accessory — are available for sale at electronics retailer Mouser and for large-volume sale through Google’s sales team. The company says that by the end of the year, it’ll expand distribution into new markets including Taiwan, Australia, New Zealand, India, Thailand, Singapore, Oman, Ghana, and the Philippines.

Coinciding with Coral’s general availability, the Coral website — which now lives at Coral.ai — has been revamped with better organization for docs and tools, testimonials, and “industry-focused” pages. Additionally, it links to a new set of examples aimed at providing solutions to common AI problems, such as image classification, object detection, pose estimation, and keyword spotting.

Lastly, Google says it’ll soon release a new version of the Mendel operating system that updates the system to the latest version of Debian (Buster), and it says it’s hard at work on updates to the Edge TPU compiler and runtime that’ll improve the model development workflow.

“We’ve received a lot of feedback over the past six months and used it to improve our platform,” wrote Coral product manager Vikram Tank. “Coral is also at the core of new applications of local AI in industries ranging from agriculture to health care to manufacturing … [It’s] already delivering impact across industries, and several of our partners are including Coral in products that require fast ML inferencing at the edge.”

For the uninitiated, the Coral Dev Board is a miniature computer featuring a removable system-on-module with one of its custom tensor processing unit (TPU) AI chips. As for the Coral USB Accelerator, it’s a USB dongle designed to speed up machine learning inference on existing Raspberry Pi and Linux systems.

TPUs are application-specific integrated circuits (ASICs) developed specifically for neural network machine learning. The first-generation design was announced in May at Google I/O, and the newest — the third generation — was detailed in May of last year.

The TPU inside the Coral Dev Board — the Edge TPU — is capable of “concurrently execut[ing]” deep feed-forward neural networks (such as convolutional networks) on high-resolution video at 30 frames per second, Google says, or a single model like MobileNet V2 at over 100 frames per second. It sends and receives data over PCIe and USB, and it taps the Google Cloud IoT Edge software stack for data management and processing.

Edge TPUs aren’t quite like the chips that accelerate algorithms in Google’s data centers — those TPUs are liquid-cooled and designed to slot into server racks, and have been used internally to power products like Google Photos, Google Cloud Vision API calls, and Google Search results. Edge TPUs, on the other hand, which measure about a fourth of a penny in size, handle calculations offline and locally, supplementing traditional microcontrollers and sensors. Moreover, they don’t train machine learning models. Instead, they run inference (prediction) with a lightweight, low-overhead version of TensorFlow that’s more power-efficient than the full-stack framework: TensorFlow Lite.

Toward that end, the Dev Board, which runs a derivative of Linux dubbed Mendel, spins up compiled and quantized TensorFlow Lite models with the aid of a quad-core NXP i.MX 8M system-on-chip paired with integrated GC7000 Lite Graphics, 1GB of LPDDR4 RAM, and 8GB of eMMC storage (expandable via microSD slot). It boasts a wireless chip that supports Wi-Fi 802.11b/g/n/ac 2.4/5GHz and Bluetooth 4.1, a 3.5mm audio jack, and a full-size HDMI 2.0a port, plus USB 2.0 and 3.0 ports, a 40-pin GPIO expansion header, and a Gigabit Ethernet port.

The Coral USB Accelerator similarly packs an Edge TPU and works at USB 2.0 speeds with any 64-bit Arm or x86 platform supported by Debian Linux. In contrast to the Dev Board, it’s got a 32-bit Arm Cortex-M0+ microprocessor running at 32MHz accompanied by 16KB of flash and 2KB of RAM.

On the subject of the camera, which is manufactured by Omnivision, it has a 1.4-micrometer sensor with an 84-degree field of view, 1/4-inch optical size, and 2.5mm focal length, and it connects to the Dev Board over a dual-lane MIPI interface. In addition to automatic exposure control, white balance, band filter, and blacklevel calibration, it features adjustable color saturation, hue, gamma, sharpness, lens correction, pixel canceling, and noise canceling.

Let’s block ads! (Why?)

Big Data – VentureBeat

AI at the edge: breakthroughs in power efficiency, learning and more (VB Live)

August 4, 2019 Big Data

Presented by Qualcomm Technologies, Inc.

AI is revolutionizing industries, products, and core capabilities by delivering dramatically enhanced experiences, but fundamental research is required to advance AI further and speed up adoption. Join this VB Live event to learn about the advances that are quickly making AI truly ubiquitous.

Access this VB Live event on demand for free right here.

Artificial intelligence is revolutionizing not just industries, but the way we think about computing in general. It’s touching everything from devices to services to appliances, and it’ll affect us in ways that we aren’t even aware of yet. There are so many ways to approach AI, and ongoing research into its potential is already changing the world.

We’re in an exciting era where we have a massive amount of intelligent wireless edge devices, says Dr. Jilei Hou, senior director of engineering at Qualcomm Technologies. As more devices become intelligent and connected, they’ll also begin to reach their limits in terms of the volume of data sent to the cloud.

In order to scale, and also address the important issues of privacy and security, AI will need to be distributed from the cloud to the wireless edge, enabling on-device processing, sensing, security, and intelligence.

“AI in terms of training and inference is largely handled in the cloud — it’s a very cloud-centric view, Hou says. “As of today, we have seen a healthy wave of AI workloads, use cases, and applications that happen on the device as well, where they are powered by power-efficient on-device AI capabilities.”

In the future, with 5G connectivity, where high speed and ultra low latency become available, we’ll see the advent of a fully distributed system where the AI workload can be flexibly deployed either in the cloud or on the device. Lifelong on-device learning capability will become mature over time, making such a fully distributed system even more powerful.

The central mission of Qualcomm’s AI research team is to make sure all these core capabilities, including perception, reasoning, and action, can be truly ubiquitous across devices and industries, including mobile, auto, IoT, and cloud. They’re developing the kind of common platform that will be fundamental to scale AI across the industry and across companies, based on three pillars: power efficiency, personalization, and efficient learning.

Power efficiency is about developing deep learning techniques to allow efficient on-device compute and achieve power and error efficiency. Personalization is about developing machine learning technology to allow adaptation to user behavior and preference. Efficient learning is the ability to train with little or no data annotation, or to handle model robustness over virtual samples.

It’s applying its generalized CN, Bayesian deep learning and optimization, deep generating models, and reinforced learning research to use cases such as deep learning for graphics, which powers computer vision, video recognition, voice UI, and also even for fingerprints. To make sure AI capability can be scaled across industries, they’re developing platform techniques and tools such as neural network quantization, compression, kernel compilation, and computed memory.

Power efficiency innovations

Power efficiency is about addressing all the computation or power consumption bottlenecks, which usually occur on both ends. Running the data through the neural model where all the math computation happens obviously contributes to power consumption. But once you load all the weights or activations in and out of the memory, the data transfer between the memory, and a compute engine can end up dominating the power consumption many times, especially in LSTM or transformer models. They’ve found that they can compound compression, quantization, compilation, and also computing memory, to achieve efficiency improvements in a compounding effect.

“With compression, quantization, and compilation, if we can achieve power efficiency improvement respectively at 3X, 4X, and 4X each, then you can just multiply them together and imagine a power efficiency improvement on the order of 50X,” Hou explains.

Quantization achievements

Quantization refers to how we can automate reduction of the position of the weights and activation while still maintaining the model’s accuracy. In general, scientists who train the models are using floating point in a 32-bit representation. However, taking a mobile device for example, a very common architecture design, you can only use 8-bit as the inference engine, but not for long.

“We already have done a good amount of research with promising results that we can transform floating point 32 into a quantized NA result, where we can still essentially keep very comparable accuracy,” Hou says. “We can achieve more than a 4X increase in the performance per watt, for a savings in memory and compute.”

Paradigm shifts in hardware architecture design

Typically memory and compute are two separate engines, and the data transfer is one of the greater performance bottlenecks — but what you can combine these two building blocks together?

“Essentially, we can embed compute engines inside memory bit cells so we can enable compute all inside of memory together in an analog compute manner,” Hou says. “If we focus on one bit ops, we can achieve potentially up to 100X power efficiency improvement.”

Want to know more?

Qualcomm continues to advance AI research to make power-efficient AI ubiquitous from device to cloud, conduct leading research and development across the entire spectrum of AI, and develop a a powerful AI platform, which is fundamental to scale AI use cases across the industry and companies. And Hou covered far more in this recent webinar.

To dive more deeply into each of these topics, learn how developer, an OEM or ISV, take advantage of some of these AI model optimization technologies, and get the details on the exciting new ecosystem and architecture advances coming your way, catch up on this VB Live event!

Access for free, on demand right here.

In this webinar, we’ll discuss:

Several research topics across the entire spectrum of AI, such as generalized CNNs and deep generative models
AI model optimization research for power efficiency, including compression, quantization, and compilation
Advances in AI research to make AI ubiquitous

Speakers:

Jilei Hou, Sr. Director, Engineering, Qualcomm Technologies, Inc.
Jack Gold, Principal Analyst, J.Gold Associates, LLC.

Let’s block ads! (Why?)

Big Data – VentureBeat

Microsoft is addressing tough paradigms on the intelligent edge with Azure Data Boxes

May 12, 2019 Big Data

This is not a rhetorical question: How do you move 2 petabytes of data to a cloud that can analyze it? It’s not practical to move it over the internet — even with an obscenely fast 10 Gbps connection, it would take 19 days. Over more common 45Mbps T3 connections, it’s effectively an impossibility, because it would take 12 years, according to Microsoft’s calculations.

This is the problem the company faced when it was developing some of its intelligent edge offerings. Its solution is Microsoft’s Azure Data Box family, a series of appliances that offer high storage capacities. Most of the Data Boxes require customers to physically ship them to Microsoft so the data can be moved to Azure, which in a sense flips the paradigm of edge computing.

Part of the idea of edge computing is, as Microsoft head of Azure IoT Sam George told VentureBeat in an interview at Build 2019, “It’s often easier to move compute to where data is than to move data where compute is.” With most of the Data Boxes, though, it’s just not feasible to do so — and thus this model is moving data to compute, not the other way around.

The amount of data a company generates varies significantly. In a conversation with VentureBeat at Microsoft Build, Kundana Palagiri, principal product manager for Microsoft Azure, said that over the course of 15 days, a company working with autonomous technologies could potentially generate up to 2PB of data. But the sweet spot for most operations is probably close to 10TB, 35TB, or 100TB over that same time period.

Some companies need to have a Data Box onsite at all times and thus would need to have a rotation of Data Boxes constantly trucking between Microsoft and the company premises. In other cases, a company will ship off their Data Box to Microsoft and be without one for a time. “Sometimes [a company] will be without it,” Palagiri said, in the case of things like “archiving, testing — various scenarios.”

In other cases, the computing and transmitting of data is actually done on the edge. There are Data Boxes for both.

From the Build 2019 stage, Microsoft showcased how grocery chain Kroger is piloting the use of Microsoft Azure services and Data Boxes in two of its stores. Announced initially in January, the program partially is using Azure-powered video analytics and AI to track inventory based on what’s physically on store shelves. Kroger points video camera arrays at its shelves, and using machine learning, it can track in real time how many units of a given product are left. It’s a tremendous amount of data, with intelligence laid over top.

The Data Box family

There are four members of the Azure Data Box family, each with unique traits designed to suit different business needs.

The most endearing, perhaps, is the Data Box Heavy. Its name is precisely descriptive: It’s a box the size of a small refrigerator, and although it’s unclear how much it weighs, it is on wheels, which tells you all you need to know about that. Inside is a bunch of hard disks totaling 1PB of capacity. For context, that’s 1,000 terabytes (TB), or 1 million gigabytes (GB). It has 4 x 40 GbE QSFP+ networking capabilities, which is extraordinarily speedy.

Microsoft is addressing tough paradigms on the intelligent edge with Azure Data Boxes

Above: Azure Data Box Heavy

Image Credit: Seth Colaner, VentureBeat

The Azure Data Box is physically smaller than its big brother Heavy, and although it offers a tenth of the storage capacity, that works out to a still quite robust 100TB. The Data Box is also designed such that the customer fills up the storage capacity and physically sends the box back to Microsoft for uploading and return. Though not as extreme as the Data Box Heavy, the networking prowess of the Data Box is significant, at 2 x 10 GbE SFP+.

Above: Azure Data Box

Image Credit: Seth Colaner, VentureBeat

Both of the big boxes offer AES 256-bit encryption with SMB/CIFS and NFS connectivity.

For a little more flexibility, customers can opt for the Azure Data Box Disk, which is actually five disks. They’re white-label drives bearing Microsoft branding. Each promises 8TB of capacity, so the total capacity per order is 40TB. Unlike the other two Data Boxes, these drives have no networking capabilities, so data has to move over a USB 3.1, SATA II, or SATA III connection. Instead of 256-AES encryption, they have 128-bit.

Above: Azure Data Box Disk

Image Credit: Seth Colaner, VentureBeat

With all of the above options, you literally mail the boxes back to Microsoft, where the company uploads or transfers the data for you. Then they’ll clear the data from storage drives and send the box back.

Edge compute

The fourth member of the family, the Azure Data Box Edge, is a departure from the rest of Data Boxes. Not only is it designed to transfer data over an internet connection to and from the cloud, it computes on the edge. This is the manifestation of the FPGA-powered sort of appliance that Microsoft announced this week at Build. Data flows from the Data Box Edge to the cloud and back, and the machine performs preprocessing, ML inferencing, and cloud-managed compute. It runs on two 10-core Intel Arria FPGAs and has 12TB of onboard storage capacity.

Above: Azure Data Box Edge

Image Credit: Seth Colaner, VentureBeat

Like the two larger Data Boxes, it has 256-bit AES encryption, as well as SMB/CIFS and NFS connectivity. For networking, Microsoft equipped it with 4 x 25 GbE SFP+ capabilities.

The other three types of Data Box, though, represent a (partial) physical metaphor for how edge computing works with and without the internet: They collect data on the edge, and that data is sent to Azure, where it undergoes training and inferencing in the cloud, and then Microsoft turns that into business insights. It’s just that in this case, instead of the data transferring from the edge to the cloud over a local internet connection, it has to move from the edge to a new physical location, and then to the cloud over a local internet connection. The Data Box Disk actually has yet another step, because it doesn’t have its own internet capabilities — once the drives arrive at a Microsoft facility, the data has to be transferred from drive to device, and then to the cloud over the internet.

These paradigms speak to the breadth of Azure services and edge computing, as well as to the practical challenges companies still face with using them — and the creative ways Microsoft is approaching potential solutions.

Let’s block ads! (Why?)

Big Data – VentureBeat

Baidu unveils open source edge computing platform and AI boards

January 11, 2019 Big Data

Baidu’s had a busy week. Fresh off of yesterday’s unveiling of Apollo 3.5, the latest generation of its self-driving platform, the Beijing company announced OpenEdge, an open source computing platform that enables developers to build edge applications “with more flexibility.” It also announced two new AI hardware development platforms: the BIE-AI-Box, a kit for in-car video analysis designed in partnership with Intel, and the BIE-AI-Board, a chipboard codeveloped with NXP that’s optimized for object classification.

“The explosive growth of IoT devices and rapid adoption of AI is fueling great demand for edge computing,” Watson Yin, vice president and general manager of Baidu Cloud, said. “Edge computing is a critical component of Baidu’s ABC (AI, Big Data and Cloud Computing) strategy.”

OpenEdge

OpenEdge is the local package component of Baidu’s commercial Baidu Intelligent Edge (BIE), which the company claims is China’s first open source edge computing platform.

The BIE platform underpinning OpenEdge offers a cloud-based management suite to manage edge nodes, edge apps, and resources such as certification, password, and program code. It supports models trained on AI frameworks such as Google’s TensorFlow and Baidu’s own PaddlePaddle, meaning that developers can train AI models on BIE and deploy them locally. Moreover, devices deployed with BIE are afforded additional features, like the ability to cache data and perform on-device processing in the event of a flaky network connection.

These tools together, Baidu says, let developers build custom edge computing systems on a range of hardware that can collect data, distribute messages, perform AI inference, and synchronize with the cloud.

“By moving the compute closer to the source of the data, it greatly reduces the latency, lowers the bandwidth usage, and ultimately brings real-time and immersive experiences to end users,” Yin said. “And by providing an open source platform, we have also greatly simplified the process for developers to create their own edge computing applications.”

BIE-AI-Box and BIE-AI-Board

Baidu and Intel’s BIE-AI-Box is, as alluded to earlier, a hardware kit custom built for analyzing the frames captured by cockpit cameras. Toward that end, it incorporates BIE technologies “specially” engineered for the purpose, and connects with cameras for road recognition, car body monitoring, driver behavior recognition, and other tasks.

As for the BIE-AI-Board, which is designed for object recognition, it’s compact enough to be embedded into cameras, drones, robots, and other hardware. Early partners have integrated it with electric vehicles to assess the health of chargers and with agricultural drones to analyze crop spectral data, Baidu says. (In the latter case, it helped to reduce pesticide use by up to 50 percent.)

Baidu’s looking to the cloud for revenue growth. It recently partnered with Nvidia to bring the chipmaker’s Volta graphics platform to Baidu Cloud, and in July 2018, it unveiled two new chips for AI workloads: the Kunlun 818-300 for machine learning model training and the Kunlun 818-100 for inference.

According to Gartner, the cloud computing market is projected to be worth $ 441 billion by 2020.

Let’s block ads! (Why?)

Big Data – VentureBeat

Dynamic Planning Meets Edge Computing To Become The New Norm

August 21, 2018 SAP

Part 6 in the “Intelligent FP&A” series

What a world we live in today! We can ask any question and get an answer almost instantly. The quality of that answer will vary as a function of the data the system is able to access, and the tool or logic used to determine the answer, but it’s still fascinating.

I’m of a vintage where the concept alone used to reside in the realm of science fiction. When you add the use of natural language processing—Siri and Alexa, for example—you might feel like a crew member on the Starship Enterprise (is my nerdiness showing?).

I bring up this observation because it is highly relevant to the topic of intelligent financial planning and analysis (iFP&A) and dynamic planning, a subset of iFP&A. Thanks to technological advances, we now have tools that permit us to view the entire financial picture of our organizations in a matter of seconds or minutes, rather than hours, days, or weeks.

The term dynamic planning refers to immediate, insightful analytics that permit FP&A professionals to assist in critical planning and decision-making and to share insights equally quickly. Today’s tools help effectively bridge the data, analytics, action, and capability gaps that exist in all organizations, and that enhance the ability to assist in effective decision-making, with the ultimate goal of improved enterprise performance.

Accelerating the transformation

Dynamic planning is expected to become the norm in business over the next four or five years. The advances and acceptance of edge computing (data processing near the source of the data) will help accelerate this transformation. Projections are that companies will integrate edge computing in 40% of their projects by 2021, compared to less than 1% in 2017.

Instant access to data increases the power to develop and execute strategies by enabling a clearer understanding of what is actually going on across the organization in real time. iFP&A professionals are thus better able to evaluate and set a strategic path while balancing short-term versus long-term drivers.

Supporting the “storytelling” approach

Dynamic planning can also assist with communications to key stakeholders. With a robust dashboard that follows IBCS standards and can be updated and revised on a “live” basis, the ability to explain what is occurring becomes easier and provides a larger audience a better understanding of the story. The requirement that FP&A professionals be outstanding “storytellers” becomes easier to achieve as the tools we use to explain and share become more relevant and useful.

Dynamic planning gives us the opportunity to:

Incorporate both internal and external sources of data
Leverage a sophisticated analytical framework
Present reports, analyses, and insights via an interactive interface
Connect to any device within the organization by using digital web-based tools
Deliver fresh insights as often as the data is refreshed (ideally in real time)

By implementing dynamic planning, we can empower our FP&A executives and teams to check on major performance metrics and identify trends in real time. They can convert data into insights faster to run what-if scenarios on the fly as new data is received.

For more on this topic, read the Dynamic Planning Series.

Upcoming events

We will be hosting several events all over the world, where we will be discussing and diving deeper into these critical issues and topics.

In addition to our Global FP&A Roundtable series, we hope to see you at Financials2018 EMEA in Prague, October 16–18, and the Financial Excellence Forum in New York City.

Follow SAP Finance online: @SAPFinance (Twitter) | LinkedIn | Facebook | YouTube

iFPA Blog CTA Dynamic Planning Meets Edge Computing To Become The New Norm

Let’s block ads! (Why?)

Digitalist Magazine

On the Edge of More Accessible IoT Innovation

April 12, 2018 TIBCO Spotfire

We’re all well-versed with the narrative that pits the cloud and edge computing against one another as competing entities.

Both have jostled for the top status as IT’s core disruptor and been positioned as a stark choice to make, depending on a business’s priorities and capabilities. Yet this ‘either/or’ conundrum is a myth worth dispelling; they are entirely different concepts. The edge — that physical space that draws computing and intelligence ever closer to the data source — becomes a delivery mechanism for the disconnected elements of the cloud. As such, they can work in synergy, rather than as replacements, making an effective hybrid that combines the edge’s agility with the sheer processing power of the central cloud. It’s why both environments are deployment options for the new breed of developers creating smarter, event-driven microservices for faster, more flexible app development.

While it was predicted that connectivity costs would reduce to such an extent that locating system intelligence in the central cloud would become the preferred option, these predictions have not come to fruition. Instead, we have seen the gradual migration of IoT-created data to the edge and with it, the natural progression of enhanced connectivity and functionality. Indeed, intelligence at the very edge of the network is not only more accessible but captured in real time, in its purest form and freshest state. This makes it most valuable for informing immediate and accurate operational decisions.

The benefits rumble on; with the compute happening directly on device, bottlenecks caused by multiple devices communicating back to a centralized core network are consigned to the past. Furthermore, security risks are minimized, with the time that data is spent in transit and vulnerable to attack significantly slashed. When analytics are added to the mix, things become more interesting, as subsets of data can be collated with analysis localized for enhanced decision making.

While the case for the edge has always been compelling, not everything has traditionally thrived there. Today’s machine learning algorithms, for example, and their requirement for huge volumes of data and computing power, have long relied on the cloud to do the heavy lifting. Yet as artificial intelligence becomes a more mainstream reality, informing daily routines from smart cars to digital personal assistants, things have changed. Attention has now turned to how it can better deliver on the network periphery and the space closer to the mobile phones, computers and other devices where the applications which commonly harness this technology run.

We have already seen the benefits play out in the smart home area. Here, deep learning capabilities at the edge of the network inform the nuances and intuitive responses of IoT-enabled digital tools that integrate and interact to provide insight as situations change. They can then feedback real-time contextual information to the homeowner or, if intruders appear, to a professional monitoring resource.

This was just the start; bringing machine learning capabilities to the edge, on device, with no connectivity requirements, and simplifying the long-standing IoT integration challenges, has wider ramifications for a host of industries and applications beyond the consumer space. Solutions that can respond to events in milliseconds represent the most cutting edge of innovation in this space, unlocking ever greater value in territories as diverse as industrial settings to the medical sector.

Here, real-time information at its most accessible will drive intelligent diagnostic capabilities on medical devices and harness machine learning to make all manner of predictions such as the patients most at risk from a hospital infection or most likely to be readmitted after discharge. At this stage, we are not privy to the full potential of AI’s role in this environment. However, a future where a medical facility can offer patients the option of receiving online medical advice from an artificial intelligence software programme is on the horizon, and promises improvements from speed and efficiency, to patient care and cost savings. Equally, strides are being made in the context of industrial settings, where data must flow between a myriad of sensors, devices, assets and machinery in the field, often in unstructured or challenging and remote conditions. Detecting anomalies at the edge device offers the kind of agility needed for predictive monitoring and mission-critical decisions with the potential to save millions, in terms of addressing equipment failures before they do damage.

Crucially, open source projects that simplify the development and deployment of microservices and IoT applications become the bedrock of this innovation. By enabling autonomous device action for a smarter edge and an era of more accessible IoT development, the potential is limitless.

Let’s block ads! (Why?)

The TIBCO Blog

Analyst Whitepaper: How Edge Computing, Serverless, and Machine Learning Will Transform the Enterprise

March 20, 2018 TIBCO Spotfire

Machine learning at the edge is becoming an integral part of modern applications. From the web to mobile to IoT, machine learning is powering a new breed of applications through natural user experiences and inbuilt intelligence, all in devices themselves that can make decisions and take actions right on the spot.

Unfortunately, today’s virtual machines and containers are too heavy of an option to be deployed at the edge. As a result, developers are moving towards a serverless compute model which uses code snippets as a unit of execution. This enables machine learning models to be packaged as functions so they can run at the edge.

The convergence of machine learning, edge computing, and serverless will become the backbone of digital transformation in industry verticals such as manufacturing, automobile, healthcare, finance, and insurance in the very near future. And TIBCO is here to help.

TIBCO has built solutions that enable developers to bring the power of machine learning and serverless computing to the edge. We now have edge computing and serverless computing platforms integrated with machine learning frameworks. Our products, like TIBCO Flogo® Enterprise and Project Mashling™, make it easy to develop and deploy intelligent applications packaged as microservices running at the edge.

To learn more about these trends, please download the analyst white paper: “How Edge Computing, Serverless, and Machine Learning Will Transform the Enterprise” today.

Let’s block ads! (Why?)

The TIBCO Blog

Tag Archives: edge

Predictive Analytics in Manufacturing: A Winning Edge

Big challenges, big rewards

Predicting return rate

Improve forecasts and maximize revenue

Improve inventory management with demand forecasting

Build your manufacturing business with analytics

Taking A Peek Over The Mountain Edge

Get 2020 vision about edge computing and 5G

This article is part of the Technology Insight series, made possible with funding from Intel.
________________________________________________________________________________________

What and why

Definition emerging

Benefits of low-latency edge computing

The edge is pervasive; 5G makes it better

The edge is still young

Google’s Coral AI edge hardware launches out of beta