Should CIOs consider Distributed Cloud?
Thu 11 Mar 2021
Will the distributed cloud model disrupt the future of cloud computing?
For most businesses, cloud computing exists in three different versions: private, public, or hybrid. However, a fourth type is making waves in the enterprise tech world: distributed cloud.
In many ways, the concept – spreading data across a number of locations rather than keeping it in a centralised data centre – is nothing new, so why is it a technology buzzword today?
What is Distributed Cloud?
Distributed cloud makes it possible for a company to run public cloud services in a number of different physical locations but manage it as a centralized service. These physical locations could include on premise, in colocation data centres, or in third-party data centres as well.
This represents a break from the traditional way of thinking about cloud computing. Physical location of data has been largely removed from the cloud conversation – that’s one of the reasons that the term ‘cloud’ was popularised in the first place.
Distributed cloud brings the physical location of data back into consideration in cloud computing.
Distributed Cloud vs. Multi-Cloud
Multi-cloud also involves spreading data and processes out to multiple clouds, across different vendors. The key difference is in a single dashboard which allows operations to manage multiple clouds and vendors at once.
When an operations team is managing a multi-cloud environment they may be challenged by differences in environments, or provider requirements and functions, which complicate operations and make tasks like user access and security difficult to manage.
Distributed cloud solves this challenge with a single control panel. With this, an operations team can manage a network that is spread across any number of providers, data centres, and geographic locations without having to switch between dashboards. Cloud management becomes fluid and streamlined, and the benefits of hybrid and multi-cloud are maximised.
Benefits of Distributed Cloud
If location hasn’t traditionally been a consideration for data storage and processing, why is it important now?
There are four main reasons why distributed cloud is garnering attention: performance, consistency, scalability, and risk.
Latency is one of the key indicators of network performance. Low latency means that data is travelling faster, speeding processing and throughput while reducing the delays that can impact the user experience. Optimising the physical location of data – to make it as close to the user as possible – reduces network latency and improves network performance.
Bandwidth is separate from latency, describing how much data can travel at one time as opposed to latency, which describes how fast it can travel. Centralised management of a distributed data network allows a company to manage its traffic and bandwidth, to ensure that the user experience isn’t only fast, it is also consistent.
It is challenging to build a network that handles existing needs but is easy to scale for growth. A distributed cloud network improves scalability because it is horizontally scalable: the capacity of each location can be increased independently of the others, but still controlled from the centralised dashboard.
Keeping data in different physical locations helps to reduce network vulnerability, making it more resilient and reducing the risk that outages or service interruptions will impact the network. The centralised network management of a distributed cloud system also supports advanced business continuity and disaster recovery.
Challenges to Distributed Cloud
The basis of distributed cloud is that network data and processes can be spread across locations managed by different cloud providers. However, multi-cloud management presents its own difficulties. In a recent survey, IT leaders reported that the top challenges associated with multi-cloud management were:
- Connectivity between providers (60%)
- Different support processes (54%)
- Varying platform services (53%)
However, as more companies leverage distributed cloud and improve the centralized dashboards and oversight tools for multi-cloud management, these challenges will become easier to master.
Future of Cloud
In fact, they have predicted that “by 2024, most cloud service platforms will provide at least some distributed cloud services that execute at the point of need,” envisioning an ATM-like network of localised data to serve low-latency application requirements.
This expectation of growth is due to a combination of factors mentioned above, but distributed cloud is also well-suited to managing large data loads associated with analytics, IoT, AI/ML, and other technologies that are being adopted and refined in a number of industries.
For all these reasons, the distributed cloud model is set to become an increasingly popular approach in the years ahead.