Kubecon + CloudNativecon India 2025 wrapped in Hyderabad last week, combining thousands of developers, platform engineers, and indigenous cloud practitioners for two days of intensive learning and collaboration. The second annual edition of India has shown not only the technical evolution of Kubernets, but the fundamental change in the way organizations think about the developer’s […]
Benchmarking vpus and gpus for media work loads
Akamai Cloud has recently introduced the support of Netint (VPUS) T1U video processing units, which make these cards powerful available at an hourly rate when using virtual cloud engines. They join NVIDIA RTX ™ 4000 ADA Generation GPUs already offered in the Cloud Akamai portfolio. To evaluate real-world performance and energy efficiency, we joined Cires21, […]
As the cloud wrapping evolved to optimize the caching of a small object
This blog was written in collaboration with Ulhas Math, whose expertise as a senior cloud wrapping architect defined the architectural evolution and the improvement of the performance described below. His guide was instrumental in addressing the scaling challenges posed by the web traffic of a small object. Cloud Wrapper launched in 2018 as Akamai’s dedicated […]
AI AGENTIC: How autonomous agents are changing the game
At the heart of many modern AI applications there are large generative models such as large language models (LLMs) or image generation models. These models are trained on huge amounts of data and may contain billions of parameters. The generative AI (Genai) has been a transformative technology, which is generating content such as text, images, […]
Generative Customization at the end: The future of e-commerce is not (simply) in the cloud
The edge computers are transforming personalization into e-commerce. The cloud is becoming the factory, not just the showroom, as personalization moves to the edge. Here’s a common scenario we see with our retail customers. Buyer Click through your homepage. Their intention is clear. You have five seconds, maybe less, to try to understand them. Only […]
Why AI systems use vector databases to improve performance and reduce costs
AI training is resource-intensive. It takes massive data sets, advanced algorithms, and specialized hardware (read: expensive GPUs) to teach a model how to understand language, image, or audio. But even after that training is complete, the inferences that run (the process of generating outputs) may be taxed in the same way. Real-time LLM question for […]
Open source & full managed: Databases working with AIVE on AKAMAI
Virtually there is no application without a database in one form or another. User data and behavior, product information, and all operating details need to be stored and made available for recovery and manipulation. There are different databases for different needs, but we can say that the most widely used are relational databases. Mysql and […]
Three ways to launch Linode Kubernets Engine Enterprise
Linode Kubernets Engine Enterprise (LKE Enterprise) launched in limited availability on June 5, and with this launch, we wanted to emphasize three ways to create the first cluster of your enterprise. This blog includes brief instructions for the use of Cloud Manager, The Linode Cli, and Terraform. If you do not already have access to […]
Because observability tools tend to fail scale
Observability is no longer just about capturing error or checking if a server is added. In modern distributed systems, it is about understanding behavior around dozens, if not thousands, of services, all that work in different environments and generate massive amounts of data. That level of complexity is exactly why choosing the right observability tool […]
Ensuring border-security navigation in integrated LLM systems
In previous parts of this series, we have explored the exciting new ways of large language models (LLMs) can integrate with APIs and act as smart agents. We have seen how LLMs and arousal interact, and we have seen how to move from basic tool descriptions to complex rules-based book design to guide them. But […]