The way applications are created has changed dramatically in just a few short years. It’s now all about rapid assembly and integration; applications built using microservices and containerized. Containers and the cloud have certainly been the catalyst for this progress and will continue to be the infrastructure on which developers create new and innovative applications.
As workloads are moving to the cloud, however, more and more businesses are realizing they need a cloud that can be rapidly adapted to their specific needs. Computing environments may be a blend of server types: On-premises, bare metal, colocated and distributed across multiple continents. No single cloud architecture can offer consistently exceptional application performance for all users.
Between hybrid, on-premises, multi-cloud, colocated, centralized and distributed clouds, it can be hard to keep all the terminology straight. What are the differences between these cloud architectures? Why use one rather than the other? In this article, we’ll focus on just one of these: The distributed cloud.
As opposed to the large public cloud where all computing is done in a central location, a distributed cloud consists of a system of servers that are not all in the same place—even though they are managed through a single cloud provider. The servers in a distributed cloud can be actually anywhere in the world. This allows for computing to be done closer to the end user, which improves the user experience in a multitude of ways.
Below are six of the key benefits of using a distributed, rather than a centralized, cloud.
Computing environments may be a blend of server types: On-premises, bare-metal, colocated and distributed across multiple continents. Cloud providers need to offer flexibility to customize a cloud for their end users and their business needs.
Many countries have specific compliance requirements for their citizens’ data and do not allow for personal/financial data to be hosted outside of their borders. Thus, having a distributed server within those countries alleviates that issue.
In a distributed cloud model, servers can be placed in a wide array of locations, including in an edge cloud. When servers are placed closer to end users as opposed to in a centralized location, data does not need to travel as far to get back to the server.
A distributed cloud can be used to deploy content delivery networks, or CDNs, to speed up content and service delivery. Heavy content (e.g. videos) can be stored and delivered by the distributed cloud closer to end users, which enhances the user experience.
Network connectivity is always a worry for cloud providers. If their system goes down, none of the user’s workloads will run. In a distributed cloud system, however, the network is at a reduced risk of failure because if the system crashes, cloud services can reside in a local subnet.
A distributed cloud can always become more distributed. More locations can be added to the network, broadening the global compute zones available within that cloud. Companies that want to reach users in specific locations around the world can use a distributed cloud to reach them.
To address the demand for versatility, a new type of cloud has emerged: The unified distributed cloud (UDC), in which workloads can run in thousands of locations anywhere on earth. Such a cloud connects to existing infrastructure around the world via one API. Businesses can enjoy the benefit of running workloads in multiple locations without needing to maintain any of the distributed infrastructure. From a single UI, users pick an existing data center to send their workloads to and, within minutes, begin deploying.
A UDC transforms heterogeneous infrastructure into a homogeneous cloud computing platform, which can then be leveraged to support the delivery of cloud services in proximity to end users. It is designed to be interoperable with any private, hybrid cloud or multi-cloud architecture. As a result, enterprises can create an optimized cloud computing strategy, taking advantage of existing and available infrastructure in the location—or locations—of their choosing. And because it sits on existing infrastructure, it doesn’t require any additional capital expenditure.
As opposed to the traditional cloud model, the UDC can change the way people think about the public cloud. While the large public cloud providers benefit from economies of scale, a distributed cloud takes advantage of the economies of locality and distribution and thus has the flexibility to handle emerging location-dependent apps, such as the new wave of internet-of-things applications.
Observability engineers are increasingly becoming essential for managing the complexity and unpredictability of modern IT systems.
In an extra #TheLongView: Everyone’s favorite single-board ARM computer, the Raspberry Pi, has a new generation coming soon.
Senser launched an AIOps platform that leverages eBPF in Linux operating systems to collect data from IT environments.