Existing & emerging trends in Cloud Computing — Part I (Primer)
Cloud Computing is referenced mostly with Public clouds and in some cases with the Private and Hybrid clouds. Recent trends in the Cloud computing is inching towards Multi-cloud, Edge computing and Distributed cloud.
This post looks into the existing trends and the emerging trends in the Cloud Computing.
Existing Trends:
Public Cloud: This cloud paradigm is popular and is a key driver to digital transformation. During COVID-19 era, most companies accelerated their journey to public clouds (AWS, Azure, GCP & others). Some of the factors that influenced this transformation were:
- Scalability & Lower Capex
- Accelerate application development with Cloud PaaS
- Extended compute for AI & ML applications with Cloud IaaS
Private Cloud: Private cloud helps organizations to develop & manage their own cloud computing infrastructure. The cost of setting up the private cloud (Open stack & others) is initially significant but eventually it reduces the TCO (Total cost of ownership) over a period of time. Private cloud is mostly preferred in these cases:
- Organizations who want greater control over their data and have strict compliance restrictions
- Always-on compute required for an organization which gives advantage on costs
- Flexibility in migration of legacy hardware
Hybrid Cloud: Hybrid cloud is a combination of on-premises, Private and Public cloud environments. Organizations prefer hybrid cloud scenarios for these cases:
- Have Critical workloads in the on-premise/private clouds and have less critical workloads in Public cloud
- Cloud bursting — Use the private cloud for predictable traffic and use Public cloud for the additional traffic during peak hours
- Move to public cloud at your own pace
Emerging Trends:
Multi-Cloud: Multi-cloud strategy provides organization the flexibility to use different services from multiple cloud providers. For example, AWS can be used for IaaS, Azure for PaaS and Google for AI/ML workloads. A hybrid cloud becomes a multi-cloud when you have more than one public provider along with the private cloud. Organizations prefer this strategy for these use cases:
- Avoid Vendor lock-in
- Guaranteed SLA
- Optimized ROI
Edge computing : Modern applications that required low latency and high bandwidth require processing close to the point where data is produced. This is the case where computing at the Edge (aka Edge computing) becomes important. Centralized cloud models do not provide the quality of experience (QOE) for these use cases.
Source: https://www.pluribusnetworks.com/solutions/distributed-cloud/
Recent advancements in mobile networking technology of 5G is driving the adoption of edge computing and vice versa. The increased bandwidth from 5G is revolutionizing the data compute at the edge. Some of the typical use cases of Edge computing are:
- Low latency applications — AR/VR, Self-driving Cars & drones
- IOT use cases
- Smart factory with Industry 4.0
Many public cloud providers are bringing their cloud to the Edge (Eg. Azure has Azure edge zones). However Edge computing seems to be a market where there will be multiple private players . In some cases, organizations can setup their own Edge compute that gives more control on the infrastructure (eg.. openstack edge computing ) or partner with providers such as Equinix, Hitachi Vantara etc..
Distributed Cloud: Distributed cloud is an extension to edge computing (or) is a trend that is becoming popular due to edge computing. Imagine a world where you have multiple edge computing sites that are available (similar to Wi-Fi hotspots) where most of the computing is performed.
Public clouds would still be used for the non latency specific workloads and you have a mix of multiple edge sites, private cloud and Public cloud in the infrastructure. Distributed cloud provides you a single control plane to get that complete visibility of your edge, private and public clouds. Distributed cloud enables location specific computing contrary to the current model of centralized cloud where location is often not a primary criteria.
Gartner predicts that most cloud service platforms will provide some form of Distributed cloud services by 2024. Some of the typical use cases of distributed cloud are:
- Improved visibility of Hybrid/multi-cloud/edge infrastructure
- IOT, AI & ML applications with use-cases similar to Edge computing
- Localized regulatory compliance
Thanks for reading through the above. Further parts of this series will focus on thoughts around Multi-cloud design & Self-service automation for existing and emerging trends in the cloud computing.
References:
https://www.liquidweb.com/blog/private-cloud/
https://www.netapp.com/hybrid-cloud/what-is-hybrid-cloud/#:~:text=Hybrid%20cloud%20refers%20to%20a,orchestration%20among%20the%20various%20platforms.
https://www.cloudindustryforum.org/content/five-reasons-why-multi-cloud-infrastructure-future-enterprise-it
https://www.ciena.com/insights/what-is/What-is-Edge-Cloud.html
https://www.cisco.com/c/en/us/solutions/enterprise-networks/edge-computing-architecture-5g.html
https://www.openstack.org/use-cases/edge-computing/
https://www.datamation.com/cloud/top-edge-computing-companies/
https://www.ibm.com/cloud/learn/distributed-cloud
https://www.gartner.com/smarterwithgartner/the-cios-guide-to-distributed-cloud/
“Hybrid cloud for Architects” by Alok Shrivastwa published by Packt