Industry executives and experts share their predictions for 2023.  Read them in this 15th annual VMblog.com series exclusive.

Predictions for Software Delivery in 2023

By Gopal Dommety and Gopinath Rebala, OpsMx

Regardless of the next disruption and distraction that arrives (as if the pandemic, new remote work models, and supply chain chaos weren't enough) one thing is certain - you'll be asked to deliver new software with more features, security, and reliability faster than ever before. While we have made huge progress in software delivery over the last few years, the next year should see some significant leaps forward. Here are my top five predictions for software delivery in 2023.

1. Developers cheer and productivity jumps as GitOps goes mainstream

We will finally get tools that enable businesses to leverage GitOps to make developers more productive, including for production software. For background, traditional CD solutions and development workflows use a "push" model, requiring a person or team to initiate a deployment into production. By contrast, GitOps uses a "pull" model, automatically comparing running code with the Git repository and initiating an update when changes are detected. Developers like this, but it can be a problem for most production environments, where organizations typically want the ability to review a pending update for security and performance issues. In 2023, "smart" GitOps tools will emerge that enable developers to work with the freedom of GitOps while also introducing some measure of control to enable organizations to protect their production environments. This has the potential to dramatically - and safely - boost developer productivity.

2. Vendors make observability across the application lifecycle actionable from a single location

In large companies, different groups take different approaches to deploying new applications and data sets, and they continue to deploy different new tools to manage their new environments. The result is tool sprawl, workflow snags and compliance gaps. But in a world of ever-growing cyberthreats, evolving privacy regulations, and economic pressure to do more with less, actionable intelligence based on observability into the flow of data is essential for data security, data management and developer productivity. Logging into each tool to obtain the necessary insight is highly inefficient and error prone to the point of undermining productivity and leading to dangerous gaps in knowledge. In 2023, we expect development and management teams to rebel against this inefficiency and demand solutions that can automatically roll up the necessary insight from all their tools into a single dashboard that enables immediate action. This will compel vendors to take advantage of advances in DevOps and DevSecOps processes to respond with new solutions that provide this actionable observability.

3. "Know Your Supplier" comes to open source as key to risk management

Know Your Supplier is an established practice in the software supply chain, but has been difficult to apply to open source solutions. It's difficult to control a volunteer environment where there may be little insight into who is involved in coding an application. Today, however, nearly every software solution, even proprietary ones, rely on some open source code, and open source software is increasingly foundational in production environments that handle highly sensitive data. In 2023, the weight of security and privacy concerns clearly falls on vendors who provide enterprise editions of open source software to vet their solutions and provide the assurance that enterprises need around security and licensing risks. Enterprises simply don't have the expertise or resources to do this themselves.

4. Advances in MLOps lead to significant increases in DevOps productivity 

The use of machine learning has grown significantly over the last few years, but so far it hasn't helped DevOps or developers much. That changes in 2023. As soaring data is generated by an ever-increasing number of processes, MLOps will be critical to preventing application teams from drowning - and 2023 will be a pivotal year as vendors begin to meet this need. Machine learning will find its way into a growing number of development operations via point solutions. For example, ML can dramatically increase the accuracy of detecting application errors, workflow anomalies and performance issues. It can help development teams determine which of the thousands of test cases they have on hand is the best to run in a specific situation. And it can enable predictions related to deployment times - helping teams determine whether application performance will meet required criteria or if the deployment needs to be held back. On the security side, ML can help enforce best practices for developer manifests and coding. 

5. Software delivery goes multi-cloud (with a little help from service mesh) 

Adoption of multi-cloud, federated Kubernetes application platforms has been held back because of security issues and the complexity of managing workloads across clusters. Now, thanks to the adoption of multi-cluster service mesh and improvements in federated architecture for Kubernetes, we expect to see the adoption of multi-cluster Kubernetes soar in 2023. Service mesh enables horizontal scaling of Kubernetes beyond a single cluster while providing increased security, both in terms of built-in secure communications and the separation of application access controls. This is exactly what's needed to support multi-cloud as a core production environment and target destination for software delivery, which is critical as companies seek to prevent vendor lock-in, meet the diverse needs of different parts of their organizations, and optimize workloads for performance, cost and security. Bonus prediction: Any company that isn't already operating on multiple cloud platforms will be doing so by the end of 2023.

It's been said all too often over the last few years that the only constant is change. Which makes the evolution of a faster, more reliable and more secure software development lifecycle one of the critical foundations of continued technology advancement. I believe 2023 will be a truly pivotal year for this journey.

##

ABOUT THE AUTHORS

Gopal Dommety is the CEO of OpsMx. Gopal is a serial entrepreneur and technology visionary. As CEO, he has built the team to scale the technology and go to market functions, and has proven product-market fit with customers like Cisco, Salesforce, Standard Chartered Bank, Juniper Networks, Albertsons, and many others. Prior to OpsMx, Gopal was the founder and CEO of N42, where he built a team of machine learning experts to address the problems companies face when running large scale virtual data centers. Gopal also was the architect behind multiple Cisco flagship products and designed Internet Protocols (RFCs) that are widely used in the Internet today. Gopal holds more than 60 patents in the area of large scale distributed systems.

Gopinath Rebala is the CTO of OpsMx, where he has overall responsibility for the machine learning and data processing architectures of OpsMx Enterprise for Spinnaker. Gopi also has a strong connection with our customers, leading design and architecture for strategic implementations. Gopi is a frequent speaker and well known leader in continuous delivery and in the Spinnaker community. Previously, Gopi was a co-founder and CTO at N42, which delivered machine learning tools for large operational systems.