随着分布式计算的普及，云计算和边缘计算等术语变得越来越普遍。这些不仅仅是激发对趋势兴趣的无意义流行语，而是推动跨行业创新的现有技术。
云(Cloud)计算和边缘计算是现代 IT 系统的关键组成部分。但这些技术究竟意味着什么？它们如何相互叠加？让我们来了解一下。
云计算简介
我们都使用 Dropbox 或OneDrive来备份我们的重要文件和数据。据说数据存储在“云”上，但这是什么意思？
简单地说，云(The Cloud)是可通过 Internet 访问的计算资源的集合。这个想法是，您可以廉价且安全地使用位于世界任何地方的工业规模硬件。
传统上，公司被迫设置和维护大型服务器以满足其内部计算需求。这会带来高昂的成本，更不用说缺乏灵活性。将应用程序迁移到云端允许公司抽象出硬件后端，根据需要请求尽可能多的资源。
完全从云端为网站和其他应用程序提供服务已成为惯例，大大简化了技术堆栈。亚马逊 AWS(Amazon AWS)和Microsoft Azure等服务是该领域的领跑者，为全球公司提供各种应用程序。
优点(Pros)
- 可扩展：(Scalable: )云(Cloud)服务可以在需要时增加，为应用程序提供灵活性，而无需硬性投资。
- 便宜：(Cheap: )服务提供商运行大型集中式服务器场比每家公司设置自己的计算机更具成本效益。这允许以比传统设置低得多的成本提供云服务。
- 简单：(Simple:)设置和管理内部数据库和API后端并非易事。将硬件抽象出来并根据需要请求计算资源更容易。
缺点(Cons)
- 网络依赖：(Network Dependent: )云服务的主要问题是完全的网络依赖。云(Cloud)服务不是网络连接较差的偏远地区的解决方案。
- 慢：(Slow: )根据云服务器的位置，通信可能需要几秒钟到几分钟的时间。在需要即时决策的应用程序（例如工业设备）中，这种延迟太大了。
- 带宽密集型：(Bandwidth Intensive: )由于云服务器负责计算和存储，需要传输大量数据。在生成大量信息（人工智能、视频录制等）的场景中，带宽要求非常昂贵。
边缘计算解释
云计算的一个问题是它对网络的依赖。对于大多数任务来说这不是问题，但有些应用程序对时间非常敏感。传输数据、在云上执行处理和接收结果的延迟很小但可以察觉。
然后是带宽的问题。涉及视频处理或 AI 算法的应用程序处理大量数据，这些数据传输到云端的成本可能很高。如果数据收集发生在网络连接受限的远程位置，则更是如此。
边缘(Edge)计算为这些问题提供了答案。它不是将数据发送到世界另一端的服务器，而是在现场或至少在附近的位置存储和处理。
这具有节省数据传输成本和消除网络延迟因素的优势。计算可以立即进行，实时给出结果，这对许多应用程序来说至关重要。
优点(Pros)
- 无延迟：(No Latency: )由于边缘计算机位于数据源，因此无需应对网络延迟。这会立即产生结果，这对于实时过程很重要。
- 减少数据传输：(Reduced Data Transmission: )边缘计算机可以处理现场的大量数据，仅将结果传输到云端。这有助于减少所需的数据传输量。
缺点(Cons)
- 比云更昂贵：(More Expensive than Cloud: )与云计算不同，边缘计算需要在每个边缘节点都有一个专用系统。根据组织中此类节点的数量，成本可能比云服务高得多。
- 复杂的设置：(Complex Setup: )使用云计算，我们只需要请求资源并构建应用程序前端。执行这些指令的硬件细节留给云服务提供商。然而，在边缘计算中，您需要构建后端，同时考虑到应用程序的需求。因此，这是一个更加复杂的过程。
云计算与(Cloud Computing Vs). 边缘计算(Edge Computing)：哪个更好？
您必须了解的第一件事是，云计算和边缘计算不是相互竞争的技术。它们不是同一问题的不同解决方案，而是完全不同的方法，解决不同的问题。
云(Cloud)计算最适合需要根据需求增加或减少的可扩展应用程序。例如， Web(Web)服务器可以在服务器高负载期间请求额外资源，从而确保无缝服务而不会产生任何永久性硬件成本。
同样，边缘计算适用于生成大量数据的实时应用程序。例如，物联网 ( IoT ) 处理连接到本地网络的智能设备。(smart devices)这些设备缺乏强大的计算机，必须依靠边缘计算机来满足其计算需求。由于涉及大量数据，对云做同样的事情会太慢而且不可行。
简而言之，云计算和边缘计算都有其用例，必须根据相关应用进行选择。
混合方法
正如我们之前所说，云计算和边缘计算不是竞争对手，而是不同问题的解决方案。这就引出了问题；它们可以一起使用吗？
答案是肯定的。许多应用程序采用混合方法，将两种技术集成以实现最高效率。例如，工业自动化机械通常连接到现场嵌入式计算机。
这台边缘计算机负责操作设备并立即执行复杂的计算。但与此同时，这台计算机也将有限的数据传输到云端，云端运行管理整个操作本身的数字框架。
通过这种方式，应用程序充分利用了两种方法的优势，依靠边缘计算进行实时计算，而其他一切都使用云计算。
哪个是最好(Best)的分布式计算技术(Computing Technology)？
边缘(Edge)计算不是云计算的升级版。这是一种不同的分布式计算方法，对于时间敏感和数据密集型应用程序非常有用。
但是，对于大多数其他应用程序来说，云计算仍然是最灵活和最具成本效益的方法。通过将存储和处理卸载到专用服务器，公司可以专注于其运营，而无需担心后端实施。
两者都是精明的 IT 专业人士必备的工具，大多数尖端设施，无论是物联网(IoT)还是其他，都利用这两种技术的组合来获得最佳结果。
Edge Computing Vs. Cloud Computing and Why It Matters
With distributed computing gaining popularity, terms lіke cloud computing and edge computing are becoming increasingly common. These aren’t just meаningless buzzwords to spark interest in a trend, but existing technologies driving innovation across industries.
Cloud computing and edge computing are critical components of the modern IT system. But what exactly do these technologies entail? And how do they stack up against each other? Let’s find out.
An Introduction to Cloud Computing
We have all used Dropbox or OneDrive to backup up our important files and data. The data is said to be stored on the “Cloud,” but what does it mean?
The Cloud, simply put, is a collection of computing resources accessible over the internet. The idea is that you can use industrial-scale hardware located anywhere in the world cheaply and securely.
Traditionally, companies were forced to set up and maintain large servers for their in-house computing needs. This incurs high costs, not to mention the lack of flexibility. Moving an application to the cloud allows a company to abstract away the hardware backend, requesting as many resources as needed.
It has become routine for websites and other applications to be served entirely from the cloud, greatly simplifying the technology stack. Services like Amazon AWS and Microsoft Azure are frontrunners in this space, powering all sorts of applications for companies worldwide.
Pros
- Scalable: Cloud services can be ramped up as and when required, providing flexibility to applications without hard investments.
- Cheap: It is more cost-effective for a service provider to run large centralized server farms than for each firm to set up its own computers. This allows cloud services to be made available at a much lower cost than traditional setups.
- Simple: Setting up and managing an in-house database and API backend is no easy undertaking. It is easier to abstract the hardware away and request computing resources as required.
Cons
- Network Dependent: The main issue with cloud services is complete network dependence. Cloud services are not a solution for remote areas with poor network connectivity.
- Slow: Depending on the location of the cloud servers, communication can take from a few seconds to several minutes. That delay is too much in applications requiring instant decisions (such as industrial equipment).
- Bandwidth Intensive: As the cloud servers are responsible for computation and storage, a lot of data needs to be transmitted. Bandwidth requirements are expensive in scenarios that generate vast information (AI, video recording, etc.).
Edge Computing Explained
A problem with cloud computing is its dependence on the network. This is not a problem for most tasks, but some applications are extremely time-sensitive. The delay in transmitting data, performing the processing on the cloud, and receiving the results is slight but perceptible.
Then there is the issue of the bandwidth. Applications involving video processing or AI algorithms work with large amounts of data, which can be expensive to transmit to the cloud. More so if the data collection occurs in a remote location, where network connectivity is limited.
Edge computing delivers an answer to these problems. Instead of sending the data to a server halfway across the world, it’s stored and processed on-site, or at least at a nearby location.
This has the advantage of saving data transmission costs and removing the factor of network latency. The computation can take place immediately, giving the results in real-time, which is vital for many applications.
Pros
- No Latency: As the edge computer is located at the source of data, there is no network latency to contend with. This gives immediate results, which is important for real-time processes.
- Reduced Data Transmission: The edge computer can process the bulk of the data at the site, transmitting only the results to the cloud. This helps reduce the volume of data transfer required.
Cons
- More Expensive than Cloud: Unlike cloud computing, edge computing requires a dedicated system at each edge node. Depending on the number of such nodes in an organization, the costs can be much higher than cloud services.
- Complex Setup: With cloud computing, all we need is to request resources and build the application frontend. The nitty-gritty of the hardware carrying out those instructions is left to the cloud service provider. In edge computing, however, you need to build the backend, taking into account the application’s needs. As a result, it is a much more involved process.
Cloud Computing Vs. Edge Computing: Which One Is Better?
The first thing that you must understand is that cloud computing and edge computing are not competing technologies. They are not different solutions to the same problem but separate approaches altogether, solving different problems.
Cloud computing is best for scalable applications that need to be ramped up or wound down according to demand. Web servers, for example, can request extra resources during periods of high server load, ensuring seamless service without incurring any permanent hardware costs.
Similarly, edge computing is suitable for real-time applications that generate a lot of data. Internet-of-Things (IoT), for example, deals with smart devices connected to a local network. These devices lack powerful computers and must rely on an edge computer for their computational needs. Doing the same thing with the cloud would be too slow and unfeasible owing to the large amounts of data involved.
In short, both cloud and edge computing have their use-cases and must be chosen according to the application in question.
The Hybrid Approach
As we have said earlier, cloud computing and edge computing are not competitors, but solutions to different problems. That begs the question; can they both be used together?
The answer is yes. Many applications take a hybrid approach, integrating both technologies for ultimate efficiency. For example, industrial automation machinery is usually connected to an on-site embedded computer.
This edge computer is responsible for operating the device and performing complex calculations without delay. But at the same time, this computer also transmits limited data to the cloud, which runs the digital framework managing the entire operation itself.
In this way, the application makes full use of the strengths of both approaches, relying on edge computing for real-time computation while using cloud computing for everything else.
Which Is the Best Distributed Computing Technology?
Edge computing is not an upgraded version of cloud computing. It is a different approach toward distributed computing that comes in handy for time-sensitive and data-intensive applications.
However, cloud computing is still the most flexible and cost-efficient approach for most other applications. By offloading storage and processing to a dedicated server, companies can focus on their operations without worrying about backend implementation.
Both are essential tools in the repertoire of a savvy IT professional, and most cutting-edge facilities, whether IoT or otherwise, leverage a combination of the two technologies to get the best results.
