What is edge computing?
August 15, 2019 - Leading venture capital research firm CBInsights has written an article detailing the growth and adoption prospects of edge computing. According to the article, cloud computing is no longer enough to instantly process and analyze data generated or about to be generated by IoT devices, connected cars and other digital platforms, and this is when edge computing can come in handy. The technology has the potential to be used in many industry sectors and play a huge role.
Here's the gist of the article:
Sometimes faster data processing is a luxury -- and sometimes it's a matter of life and death.
Self-driving cars, for example, are essentially high-performance computers on wheels that collect data through a vast array of sensors. In order for these vehicles to operate safely and reliably, they need to react immediately to their surroundings. Any delay in processing speed could be fatal. While data processing for connected devices now takes place primarily in the cloud, it can take seconds to transfer data back and forth between centralized servers. That time span is far too long.
Edge computing, on the other hand, makes it possible for self-driving cars to process data more quickly. This technology allows connected devices to process data that is created at the "edge," which is defined as inside the device or much closer to the device itself.
It is estimated that by 2020, each person will generate an average of 1.5GB of data per day. As more and more devices connect to the Internet and generate data, the cloud may not be able to fully process that data -- especially in certain usage scenarios where data needs to be processed very quickly.
Edge computing is an alternative to the cloud, and it will likely be used for much more than driverless cars in the future.
Some tech giants, including Amazon, Microsoft and Google, are exploring "edge computing" technologies that could spark the next big computing race. While AmazonWebServices (AWS), the Amazon cloud service, still dominates the public **** cloud space, it remains to be seen who will be the leader in this emerging edge computing space.
In this article, we'll take a deeper look at what edge computing is, the benefits associated with the technology, and how it's being used in a variety of industries.
A changing computing landscape
Before we can understand edge computing, we must take a look at how its predecessor, cloud computing, paved the way for Internet of Things (IoT) devices that are spread across the globe.
Cloud computing empowers a connected world
From wearables to connected kitchen appliances, connected devices are arguably everywhere. The global IoT market is estimated to exceed $1.7 trillion by 2019, more than tripling from $486 billion in 2013.
As a result, cloud computing - the process by which many smart devices connect to the Internet to function - has become an increasingly mainstream trend.
Cloud computing allows companies to store and process data (and other computing tasks) outside of their own physical hardware, via a network of remote servers, commonly referred to as "the cloud.
For example, you can choose to use Apple's iCloud cloud service to back up your smartphone, and then you can retrieve the data on your smartphone from another Internet-connected device, such as your desktop computer, by logging into your account and connecting to the cloud. Your information is no longer limited by the capacity of your smartphone or desktop's internal hard drive.
This is just one of many cloud computing use cases. Another example is accessing a variety of complete applications through a Web-based or mobile browser. As cloud computing has grown in popularity, it has attracted big tech companies like Amazon Google, Microsoft and IBM into the fold. According to a survey conducted in 2018 by private cloud management company RightScale, Amazon AWS and Microsoft Azure are first and second among the major public **** cloud providers.
Illustration: more businesses are running applications on the public **** cloud
But centralized cloud computing isn't right for all applications and use cases. Edge computing, on the other hand, can provide solutions in areas that may be difficult to address with traditional cloud infrastructure.
The shift to edge computing
In a future where we're inundated with data everywhere, billions of devices will be connected to the Internet, so faster and more reliable data processing will become critical.
In recent years, the consolidated and centralized nature of cloud computing has proved to be cost-effective and flexible, but the rise of the Internet of Things and mobile computing has put a considerable strain on network bandwidth.
Eventually, not all smart devices need to utilize the cloud to function. In some cases, this back-and-forth transfer of data can -- and should -- be avoided.
This has led to the emergence of edge computing.
The global edge computing market is expected to reach $6.72 billion by 2022, according to CBInsights' Market Size Quantification Tool. While it's an emerging field, edge computing is likely to operate more efficiently in some areas covered by cloud computing.
Edge computing enables data to be processed at the nearest end, such as a motor, pump, generator or other sensor, reducing the need to transfer data back and forth between clouds.
Market research firm IDC says edge computing is described as "a mesh network of miniature data centers that process or store critical data locally and push all incoming data to a central data center or cloud repository that covers less than 100 square feet."
For example, a train may contain sensors that can immediately provide information about the status of its engine. In edge computing, the sensor data doesn't need to be transmitted to a data center on the train or in the cloud to see if something is affecting the engine's operation.
Localized data processing and storage puts less strain on the computing network. As less data is sent to the cloud, the likelihood of latency -- the delay in data processing caused by interactions between the cloud and IoT devices -- is reduced.
This also puts more tasks on the shoulders of hardware based on edge computing technologies, which contain sensors for collecting data and CPUs or GPUs for processing data in connected devices.
With the rise of edge computing, it's also important to understand another technology involved in edge devices: fog computing.
Edge computing specifically refers to computing processes that take place at or near the "edge" of the network, while fog computing refers to the network connection between the edge device and the cloud.
In other words, fog computing brings the cloud closer to the edge of the network; thus, according to OpenFog, "fog computing always uses edge computing, rather than edge computing always using fog computing."
Back to our train scenario: sensors can collect data, but they can't immediately act on it. For example, if a train engineer wants to understand how the train's wheels and brakes are operating, he can use historically accumulated sensor data to predict whether parts need repair.
In this case, the data is processed using edge computing, but it's not always instantaneous (unlike determining the state of an engine). With fog computing, on the other hand, short-term analytics can be realized at a given point in time without the need to return to the central cloud altogether.
Figure: cloud, fog, and edge computing
So it's important to keep in mind that while edge computing complements cloud computing and operates very closely with fog computing, it's by no means a replacement for either.
Benefits of Edge Computing
While edge computing is an emerging field, it has some obvious benefits, including:
-Real-time or faster data processing and analytics: data is processed closer to the source of the data, rather than in an external data center or cloud, so latency times can be reduced.
-Lower costs: organizations spend less on data management solutions for local devices than on cloud and data center networks.
-Less network traffic: as the number of IoT devices increases, data generation continues to grow at a record pace. As a result, network bandwidth becomes more limited, overwhelming the cloud and creating greater data bottlenecks.
-Higher application runtime efficiency: as lag is reduced, applications are able to run more efficiently at faster speeds.
Weakening the role of the cloud also reduces the likelihood of a single point of failure.
For example, if a company uses a centralized cloud to store its data, and the cloud goes down, the data will be inaccessible until the problem is fixed - and the company could suffer a serious loss of business as a result.
In 2016, the Salesforce website's North American 14 site (aka NA14) was down for more than 24 hours. Customers were unable to access user data, from phone numbers to emails and more, and business operations were severely disrupted.
Since then, Salesforce has moved its IoT cloud to Amazon's AWS, but the downtime highlights one of the major drawbacks of relying solely on the cloud.
Reducing reliance on the cloud also means that certain devices can be stabilized to run offline. This can come in especially handy in areas where Internet connectivity is limited - whether in specific areas where there's a severe lack of Internet service or in remote areas such as oilfields that are often inaccessible.
Another key benefit of edge computing relates to security and compliance. This is especially important as governments become increasingly concerned about how businesses utilize consumer data.
The General Data Protection Regulation (GDPR), recently implemented by the European Union (EU), is an example. The regulation aims to protect personally identifiable information from data misuse.
Because edge devices are able to collect and process data locally, data does not have to be transmitted to the cloud. As a result, sensitive information doesn't need to go through the network, so if the cloud is attacked by a cyberattack, the impact won't be as severe.
Edge computing also enables interoperability between emerging connected devices and older, "legacy" devices. It takes the communication protocols used by legacy systems and "translates them into a language that modern connected devices can understand. This means that legacy industrial equipment can connect seamlessly and efficiently to modern IoT platforms.
The State of Edge Computing
Today, the edge computing market is still in its early stages of development. But it seems to be getting a lot of attention as more and more devices are connected to the network.
The same companies that dominate the cloud computing market (Amazon, Google, and Microsoft) are becoming leaders in edge computing.
Last year, Amazon got ahead of the industry by entering the edge computing space with AWSGreengrass. The service extends AWS to devices so they can "process the data they generate locally while still using the cloud for management, data analysis and persistent storage."
Microsoft is also making some big moves in this space. The company plans to spend $5 billion on the Internet of Things over the next four years, including edge computing projects.
Microsoft unveiled its AzureIoTEdge solution, which "extends cloud analytics to edge devices" with support for offline use. The company also wants to focus on AI applications at the edge.
Google isn't far behind. It announced two new products earlier this month meant to help improve the development of edge-connected devices. They are the hardware chip EdgeTPU and the software stack CloudIoTEdge.
Google said, "CloudIoTEdge extends the powerful data processing and machine learning capabilities of the Google Cloud to billions of edge devices, such as robotic arms, wind turbines, and oil drilling towers, so they are able to manipulate data from data from their sensors to operate in real time and make predictions about outcomes locally."
The three tech giants aren't the only ones interested in dipping their toes in the water, however.
As more and more connected devices emerge, many players in the emerging ecosystem are developing software and technology to help edge computing take off.
Over the next four years, Hewlett Packard Enterprise will invest $4 billion in edge computing. The company's EdgelineConvergedEdgeSystems system is targeted at industrial partners who want data center-grade computing power and often operate in remote locations.
Its system promises to provide industrial operations, such as oil rigs, factories or copper mines, with insight from connected devices without relying on sending data to the cloud or data center.
Other major contenders in the emerging edge computing space include ScaleComputing, Vertiv, Huawei, Fujitsu and Nokia.
Artificial intelligence chip maker NVIDIA launched JetsonTX2, an AI computing platform for edge devices, in 2017. Its predecessor was JetsonTX1, which claimed to "redefine what's possible in extending advanced AI from the cloud to the edge."
Many notable companies are also investing in edge computing, including General Electric, Intel, Dell, IBM, Cisco, Hewlett Packard Enterprise, Microsoft, SAPSE, and ATT.
For example, in the private equity market, both Dell and Intel have invested in Foghorn, which provides edge intelligence for commercial and industrial IoT applications. Dell also participated in a seed round of funding for IOTech, an IoT edge platform.
Many of the companies mentioned above, including Cisco, Dell and Microsoft, have also joined together to form the OpenFog Alliance. The organization's goal is to standardize the use of this technology.
Edge computing across industries
As the price of sensors and the cost of computing continue to fall, more "stuff" will be connected to the Internet.
As more connected devices become available, edge computing will be used more and more in a variety of industries, especially in areas where cloud computing is inefficient.
We're already starting to see the technology make an impact in a number of different industry sectors.
"As we bring the power of the cloud down to the device (i.e., the edge), we can bring the ability to respond, analyze, and act in real time, especially in areas where there is limited or no access to networks It's still in the early stages of development, but we're starting to see these new capabilities being applied to solve some of the major challenges on a global scale. " -- Kevin Scott, CTO, Microsoft
From self-driving cars to agriculture, here are a few industries that will benefit from the potential of edge computing.
Transportation
One of the most obvious potential applications of edge computing technology is transportation - more specifically, driverless cars.
Self-driving cars are equipped with a wide variety of sensors, from cameras to radar to laser systems, to help run the vehicle.
As mentioned earlier, these self-driving cars can utilize edge computing to process data closer to the vehicle through these sensors, which in turn minimizes the system's response time while driving. While driverless cars aren't yet a mainstream trend, companies are saving for a rainy day.
Earlier this year, the Automotive Edge Computing Consortium (AECC) announced that it would be launching programs focused on connected car solutions.
"Connected cars are rapidly expanding from luxury models and premium brands to high-volume mid-range models. The automotive industry will soon reach a tipping point when the amount of data generated by automobiles will exceed existing cloud, computing and communications infrastructure resources." --Kenichi Murata, chairman and president of AECC
The alliance's members include companies such as DENSOCorporation, Toyota, ATT, Ericsson, Intel and others.
But it's not just self-driving cars that generate tons of data and need to process it in real time. So do airplanes, trains, and other modes of transportation - whether they have human drivers or not.
Aircraft maker Bombardier's CSeries planes, for example, are equipped with a plethora of sensors to quickly detect engine performance problems. During a 12-hour flight, the airplane generates as much as 844 terabytes of data. Edge computing supports real-time processing of the data so the company can proactively address engine problems.
Healthcare
Today, people are increasingly comfortable wearing fitness tracking devices, blood glucose monitors, smartwatches and other wearables that monitor their health.
But real-time analytics may be essential to truly benefit from the massive amounts of data collected -- many wearables connect directly to the cloud, but there are others that support offline operation.
Some wearable health monitors can locally analyze pulse data or sleep patterns without connecting to the cloud. Doctors can then assess the patient on the spot and provide instant feedback on the patient's health.
But the potential of edge computing in healthcare is far from limited to wearables.
Think about the benefits that rapid data processing can bring to remote patient monitoring, inpatient care, and healthcare management in hospitals and clinics.
Physicians and clinicians would be able to provide faster and better care to patients, while the health data generated by patients has an additional layer of security. The average hospital bed has more than 20 connected devices, which generates a lot of data. The processing of this data will happen directly closer to the edge, rather than sending confidential data to the cloud, thus avoiding the risk of inappropriate access to the data.
As mentioned earlier, localized data processing means that widespread cloud or network outages won't affect business operations. Even if cloud operations are disrupted, these hospital sensors can function independently.
Manufacturing
Smart manufacturing promises to gain insight from the large number of sensors deployed in modern factories.
Because of the ability to reduce lag, edge computing may enable manufacturing processes to respond and change more quickly, with the ability to apply insights and real-time actions derived from data analysis in real time. This could include shutting down machines before they overheat.
A factory could use two robots to perform the same task, both equipped with sensors and connected to an edge device. The edge device can predict whether one of the robots will fail to operate by running a machine learning model.
If the edge device determines that the robot is likely to fail, it triggers an action to stop or slow down the robot's operation. This would allow the plant to assess potential failures in real time.
Robots could also become more self-sufficient and responsive if they can process data on their own.
Edge computing should support faster access to more insights from big data, as well as support the application of more machine learning techniques to business operations.
The ultimate goal is to unlock the immense value of massive amounts of data generated in real time, preventing safety hazards and reducing disruptions to machine operations on the factory floor.
Agriculture and smart farms
Edge computing is ideally suited for agriculture, as farms are often in remote locations and harsh environments that can have issues with bandwidth and network connectivity.
Right now, smart farms that want to improve their network connectivity need to invest in expensive fiber, microwave connections, or have a satellite that operates 24/7; edge computing is a suitable and cost-effective alternative.
Smart farms can use edge computing to monitor temperature and equipment performance, as well as to automatically slow down or shut down various devices, such as overheating pumps.
Energy and grid control
Edge computing may be particularly effective across the energy industry, especially in monitoring the security of oil and gas facilities.
Pressure and humidity sensors, for example, should be closely monitored and cannot afford to make mistakes in connectivity, especially given that most of these sensors are located in remote areas. If an anomaly - such as an overheated fuel line - goes unnoticed in time, there could be a catastrophic explosion.
Another benefit of edge computing is the ability to detect equipment failures in real time. With grid control, sensors can monitor the energy generated by everything from electric cars to wind farms, helping to make decisions accordingly to reduce costs and improve energy productivity.
Other Industry Sectors
Other industries that can utilize edge computing technology include the financial and retail industries. Both industries use large customer and back-end datasets to provide everything from stock picking information to in-store clothing placement, and could benefit from a reduced reliance on cloud computing.
Retail can use edge computing applications to enhance the customer experience. With so many retailers today working to improve the in-store experience, optimizing the way data is collected and analyzed definitely makes sense for them -- especially given that many retailers are already experimenting with connected smart displays.
In addition, many use point-of-sale data generated by in-store tablets, which is transferred to the cloud or data center. With edge computing, data can be analyzed locally, reducing the risk of sensitive data leakage.
Summary
From wearables to cars to robots, IoT devices are showing increasing momentum.
As we move toward a more connected ecosystem, data generation will continue to skyrocket, especially as 5G technology takes off, further accelerating network connectivity. While central clouds or data centers have traditionally been the preferred choice for data management, processing, and storage, both solutions have limitations. Edge computing can act as an alternative solution, but since the technology is still in its infancy, it is difficult to predict its future.
Device-capability challenges -- including the ability to develop software and hardware that can handle computing tasks that are shunted from the cloud -- are likely to emerge. The ability to teach machines to switch between computing tasks that can be performed at the edge and those that need to be performed in the cloud is also a challenge.
Even so, as edge computing becomes more adopted, there will be more opportunities for organizations to test and deploy the technology in a variety of areas.
Some use cases may demonstrate the value of edge computing more than others, but overall, the potential impact of the technology on our entire connected ecosystem could be dramatic.
Link to original article:/hello_zybwl/article/details/89219832
mec trials?
mec is edge computing technology (MobileEdgeComputing), mec is the trial is the mec's platform traffic channeling, billing and other native features and 5G cloud function of the combination of the test done by the second phase of the 5G technology trials focus on content.
mec is a key technology to support operators to carry out 5G network transformation to meet the needs of high-definition video, VR/AR, industrial Internet, car networking and other business development. Along with the formation of 5G core network SBA architecture and the rapid development of cloud computing, the current technical form of edge computing has been formed. Through mec in the country's key cities of the full deployment, in order to realize the 5G can emerge.
What does edge computing mean?
Edge computing refers to algorithms that are bounded by the "edge" of the network, such as computing inside smart gateways and cameras. But it's not practical to store or compute all of the data collected by these devices -- there's too much interfering or redundant information, which can be counterproductive if not handled properly.
Take the Hippo forest fire monitoring system as an example. The built-in pyrotechnic identification processor transmits tb of video data, but only a few megabytes of that data are of real value in terms of arousing suspicion or illegal activity, and edge computing is well suited to handle the target data of interest.
Additionally, edge computing reduces congestion on network traffic compared to cloud computing, leaving room for more mission-critical execution.