Richard Nuñez

🚨 IBM just finalized its $11B acquisition of Confluent. What does this mean for the future of your data streaming architecture? When a tech giant spends $11 billion, they are looking for a rapid return on investment. Historically (think back to the RedHat acquisition), that has meant product bundling, gatekeeping, and inevitable price hikes. If your AI and data strategies rely on real-time streaming, are you prepared for the "Streaming Tax" to get even heavier? We've just published a must-read breakdown of the acquisition and what it means for Apache Kafka® users. Here is what you need to know: 💸 The "Streaming Tax" is Scaling: Running default Kafka in the cloud comes with massive cross-AZ networking and replication costs. As your data grows, this becomes unsustainable. 🔓 Data Ownership is the Only AI Strategy: Proprietary forks of Kafka lock your data in. Aiven’s philosophy remains 100% upstream open source. No gatekeeping, no vendor lock-in. 🚀 A "Diskless" Future: Aiven engineers and Apache Kafka maintainers are driving KIP-1150 (Diskless Topics). By decoupling compute from storage, you can eliminate the replication tax and potentially reduce infrastructure costs by up to 97%! 🤝 Zero-Risk Migration: Worried about the "double bubble" of paying for two platforms during a migration? Aiven just launched a Confluent Migration Program offering 2 to 6 months of free service to make the transition cost-neutral. If you are concerned about your Kafka roadmap being "IBM-ified," or if you just want to drastically lower your Total Cost of Ownership with a truly open architecture, this is worth your time. 📖 Read the full breakdown and see how to escape the streaming tax: https://lnkd.in/dJCuFZeh What are your thoughts on the acquisition? Are you sticking with Confluent, or looking at purely open-source alternatives? #ApacheKafka #DataStreaming #IBM #Confluent #OpenSource #Aiven #CloudArchitecture #TechNews #DataEngineering #AI Aiven

3

Juha Holkkola, our chief technologist, talks with The Tolly Group about how FusionLayer #DDI addresses legacy hardware challenges in hypervisor environments by leveraging a purpose-built platform for distributed setups. #networkautomation https://lnkd.in/dyiUwme2

9

2 Comments

OCI’s supercomputer streamlines GPU management for NVIDIA GB200 NVL72 deployments. See how dedicated OCI APIs automate provisioning, clustering, and orchestration for large-scale AI workloads—unlocking new levels of speed, scalability, and operational simplicity: https://lnkd.in/eq5TVvCk

88

Is 2026 the year AI agents finally stop working in isolation? According to the latest 2026 #predictions from Kyle Campos of CloudBolt Software, we are hitting a turning point where technology silos don't just break down—they converge. Kyle predicts a shift from scattered experimentation to intentional orchestration. The winners in 2026 won't be those chasing the newest AI capability, but those building the "connective tissue" between them. Read the full prediction article here: https://lnkd.in/gqk8-UJZ #VMblog #TechPredictions #AI #CloudInfrastructure #Kubernetes #FinOps #CloudBolt #2026Trends

4

#PlanetNix 2026 ❄️ Join Priya Ananthasankar, Principal Software Engineer at Microsoft, as she shares real-world lessons from building and scaling distributed systems, and how separating runtime isolation from tooling changes the game for modern platform. Managed services face unsustainable container bloat: images balloon to multi-GB artifacts with growth. Traditional "shift-left" approaches force an impossible tradeoff: bloated monolithic images or fragmented specialized images. Runtime package layering resolves this: containers provide isolation and security while remaining small; Flox environments deliver tool freshness without compromising security. This separation achieves stability through pinned packages and freshness through Nix/Flox. See you in Pasadena? 😎 Sign up today 👉 https://buff.ly/YnqQ20u

14

1 Comment

If architected correctly, the #MCP gateway will determine which MCP to leverage to make an API call, which formats the request to its associated platform. The #API moves data from its deployed (or prepared) state to its activated state. Bob Walczak, CEO and founder at MadConnect shares more: https://lnkd.in/dWyA2cap #AI #agenticAI #martech #salestech #content #marketing

6

Insightful article by Wind River CTO Paul Miller on rethinking cloud infrastructure in the post-VMware era: https://ow.ly/16JH30sRjba #CloudFreedom #VMwareAlternative #CloudPlatform #WeAreWindRiver

4

Unlock new levels of performance with a Hammerspace Tier 0 architecture on OCI. OCI has built a high-performance architecture for HPC and AI workloads. But I/O bottlenecks, even in the cloud, cause GPUs to sit idle, waiting for data. OCI recently validated the performance of the Hammerspace Tier 0 architecture on bare metal shapes and demonstrated: - 6x improvement in read bandwidth vs networked storage - 3x improvement in write bandwidth vs networked storage Want to learn how you can use this architecture to accelerate AI pipelines on OCI? Meet with us at Oracle AI World: https://okt.to/Zd265x Explore the full results of the FIO Benchmarks here: https://okt.to/UoTabR #OracleAIWorld25 #Hammerspace Raj Sharma Jon Flynn

19