The evolution of cloud-native technology has been nothing short of revolutionary. As we step into 2024, the cornerstone of cloud-native technology, Kubernetes, will turn ten years old. It continues to solidify its position and is anticipated to reach USD 5575.67 million by 2028, with a forecasted Compound Annual Growth Rate (CAGR) of 18.51% in the coming years, as reported by Industry Research Biz.
The Cloud Native landscape continues to encompass both micro-trends and IT macro-trends, influencing and transforming the way businesses operate and deliver value to their customers.
As we at Percona wind down 2023 and look into what the next year holds, our attention is drawn to the cloud-native landscape and how it is maturing, growing, and evolving.
KubeCon NA 2023 recap
The theme for KubeCon NA was very clear — AI and Large Language Models (LLMs). Keynotes were focused on how Kubernetes and Cloud Native help businesses embrace the new AI era. And it is understandable, as Kubernetes slowly becomes what it is intended to be – the Platform.
The field of Platform Engineering has witnessed significant advancements, as evidenced by the publication of the CNCF platform whitepaper and the introduction of a dedicated Platform Engineering day at the upcoming KubeCon event. At Percona, we observe a growing trend among companies utilizing Kubernetes as a means to offer services to their teams, fostering expedited software delivery and driving business growth.
Declarative GitOps management, with ArgoCD and Flux, is the community way of adding orchestration on top of orchestration. In our conversations with developers and engineers during the conference, we confirmed the CNCF GItOps Microsurvey data – 91% are already using GitOps.
According to the Dynatrace Kubernetes in the Wild 2023 report, a significant 71% (with 48% year-over-year growth!) of respondents are currently utilizing databases in Kubernetes (k8s). This finding aligns with the observations made at the Data on Kubernetes (DoK) day, where discussions surrounding this topic transitioned from niche, tech-oriented conversations a year ago to more widespread, enterprise-level interest in adopting diverse use cases. These indicators suggest that the adoption of databases on k8s is in its early stages and is likely to continue growing in the future.
Predictions
Multi-cloud is a requirement
While this wave has been building for years, in 2024, we expect it to peak. According to a 2023 Forrester survey commissioned by Hashicorp, 61% of respondents had implemented, were expanding, or were upgrading their multi-cloud strategy. We expect that number to rise higher in 2024.
Nearly every vendor at Kubecon and every person we spoke to had some form of a multi-cloud requirement or strategy. Sometimes, this comes from necessity through acquisition or mergers. Oftentimes, it is a pillar of modern infrastructure strategy to avoid cloud vendor lock-in. At this point, it is ubiquitous, and if it is not part of your strategy, you are falling behind.
The business value of adopting this strategy is multi-fold:
- Freedom from vendor lock-in, which leads to increased negotiating power
- Agility in capitalizing on cloud-vendor advancements to innovate faster
- Increased application and database architecture RPO and RTO
- Adhering to security and governance requirements of customers
Percona’s Operators for MySQL, MongoDB, and PostgreSQL are designed with this value in mind. We want adopters of our technology to be able to deploy their critical open source databases and applications across any public or private cloud environment. All of the database automation for running a highly available, resilient, and secure database is built into the operator to simplify the operation and management of your clusters.
Simplify and secure
Looking through various State of Kubernetes reports (VMWare, RedHat, SpectroCloud), it becomes clear that Complexity and Security are the top concerns for platform engineering teams.
Simplification might come from different angles. Deployment is mostly solved already, whereas management and operations are still not. We expect to see various tooling and core patches to automate scaling, upgrades, migrations, troubleshooting, and more.
Operators are an integral part of solving the complexity problem, where they take away the need for learning k8s primitives and application configuration internals. They also remove toil and allow engineers to focus on application development vs platform engineering work. Not only will new operators appear, but existing operators will mature and provide capabilities that meet or exceed managed services that users can get on public clouds.
The latest report on Kubernetes adoption, security, and market trends in 2023 revealed that 67% reported delaying or slowing down deployment due to Kubernetes security concerns. Additionally, 37% of respondents experienced revenue or customer loss due to a container/Kubernetes security incident.
Considering the open source software vulnerability as one of the top concerns and the rapid increase in supply chain attacks (the SolarWinds attack and vulnerabilities like Log4Shell and Spring4Shell), along with container and Kubernetes strategies, there’s a growing emphasis on cybersecurity and operational understanding in development.
Another significant issue within security concerns is the escalating complexity of modern systems, especially in platforms like Kubernetes, which highlights the need for unified threat models and scanning tools to address vulnerabilities. Standardization and collaboration are key to sharing common knowledge and patterns across teams and infrastructures. Creating repositories for memory-safe patterns in cloud systems to improve overall security.
A majority of RedHat’s security research respondents have a DevSecOps initiative underway. Most organizations are embracing DevSecOps, a term that covers processes and tooling enabling security to be integrated into the application development life cycle rather than treated as a separate process. However, 17% of organizations operate security separately from DevOps, lacking any DevSecOps initiatives. Consequently, they might miss out on the benefits of integrating security into the SDLC, such as enhanced efficiency, speed, and quality in software delivery.
AI and MLOps
Kubernetes has become a new web server for many production AI workloads, focusing on facilitating the development and deployment of AI applications, including model training. The newly formed Open Source AI Alliance, led by META and IBM, promises to support open-source AI. It comprises numerous organizations from various sectors, including software, hardware, nonprofit, public, and academic. The goal is to collaboratively develop tools and programs facilitating open development and run scalable and distributed training jobs for popular frameworks such as PyTorch, TensorFlow, MPI, MXNet, PaddlePaddle, and XGBoost.
While integrating AI and machine learning into cloud-native architectures, there’s an increasing demand from users for AI to be open and collaborative. The emergence of trends stemming from ‘AI Advancements and Ethical Concerns’ cannot be ignored.
Addressing ethical concerns and biases will necessitate the implementation of transparent AI frameworks and ethical guidelines during application development. Customers will increasingly prioritize AI efficiency and education to tackle legal and ethical concerns. This marks the end of an era of chaos, paving the way for efficiency gains, quicker innovation, and standardized practices.
Conclusion
At Percona, we prioritize staying ahead of market trends by adhering to industry best practices and leveraging our team’s expertise.
We’ve always made sure to focus on security in our software development, and weaving multi-cloud deployment into our products has been a crucial part of our strategy. Our commitment to open source software drives us to take additional precautions, ensuring operational security through best practices and principles, such as of least privilege, security in layers, and separation of roles/responsibilities through policy and software controls. And with multi-cloud in mind, we consistently incorporate new sharding functionalities into our roadmaps, such as the upcoming Shard-per-location support in the Percona Operator for MongoDB.
At the same time, we are not hesitating to rock the cloud-native community by incorporating top-notch features to address any new rising trends. You mentioned ‘More Simple Kubernetes’? Well, here we are – with storage autoscaling for databases in Kubernetes, slated for release in Q1, 2024 after a year of hard work. This fully automated scaling and tuning will enable a serverless-like experience in our Operators and Everest. Developers will receive the endpoint without needing to consider resources and tuning at all. It’s worry-free and doesn’t require human intervention.
Finally, the rising popularity of generative AI and engines like OpenAI or Bard has prompted our team to bring vector-handling capabilities to Percona-powered database software by adding support for the pgvector extension.
Our team always focuses on innovation to accelerate progress for everyone, and we will continue to push the boundaries further for our community and the rest of the world.
The Percona Kubernetes Operators automate the creation, alteration, or deletion of members in your Percona Distribution for MySQL, MongoDB, or PostgreSQL environment.
Learn More About Percona Kubernetes Operators
