Affordable servers with the best price-performance ratio on the market

Start

Perfect balance of processing power, memory and storage

The high performance backbone of your mission-critical infrastructure

Core

For mission-critical data, fast storage, backup and streaming

Store

Dedicated GPU power with reliable performance and stability

Host your Hardware in our secured French datacenters

Dedirack

60 locations worldwide, starting at €4,99/month

Dedibox VPS

Dedibox - dedicated servers

Fully dedicated bare metal servers with native cloud integration, at the best price

Aluminium

Powerful, balanced and reliable servers for production-grade applications

Beryllium

Powerful dedicated server designed to handle high-workload applications

Iridium

Designed with huge local storage to keep, back up, and protect your data

Lithium

Power and stability of dedicated GPU hardware integrated into the Scaleway ecosystem

Titanium

Elastic Metal - bare metal cloud

Enjoy the Mac mini experience with great simplicity

Mac mini M1

Perform your daily tasks with speed and efficiency

Mac mini M2

Realize your most ambitious projects thanks to a new level of power

Mac mini M2 Pro

Apple

Dedibox and Bare Metal

Dedicated vCPU for the most demanding workloads (x86)

Production-Optimized

Secure, scalable VMs, equipped for high memory and compute demands (x86)

Workload-Optimized

Highly reliable and priced affordably Instances with shared vCPUs (x86 and ARM)

Cost-Optimized

A tiny instance to test and host your personal projects (x86)

Learning

Virtual Instances

Maximize your AI infrastructures with a versatile Instance

L4 GPU Instance

Universal Instance, faster than L4 and cheaper than H100 PCIe

L40S GPU Instance

Accelerate your model training with the most high-end AI chip

H100 PCIe GPU Instance

Dedicated NVIDIA® RTX 3070 with the best price/performance ratio

GPU 3070 Instances

Dedicated Tesla P100s for all your Machine Learning & Artificial Intelligence needs.

Render GPU Instances

Experience an easy way to run your code on the cloud

Serverless Functions

Easily run containers on the cloud with a single command

Serverless Containers

Serverless Jobs

Serverless

Kubernetes exclusively for Scaleway products and resources

Kubernetes Kapsule

Multi-cloud Kubernetes for Scaleway and external providers resources

Kubernetes Kosmos

Container Registry

Containers

Compute

Build the next Foundation Model with one of the fastest and most energy-efficient supercomputers in the world

Custom-built Clusters

Rent a GPU-cluster from 32 to more than a thousand GPUs to speed up distributed training

On Demand Cluster

Clusters

Deploy AI models in a dedicated inference infrastructure. Get tailored security and predictable throughput

Managed Inference

Consume AI models instantly via a simple API call. All hosted in Europe

Generative APIs

Model-as-a-service

Accelerate the next generation of AI-enabled applications with the universal L40S GPU Instance, faster than L4 and cheaper than H100 PCIe

Maximize your AI infrastructure's potential with a versatile and cost-effective GPU Instance

Dedicated Tesla P100s for all your Machine Learning & Artificial Intelligence needs

Render GPU Instance

GPU Instances

Amazon S3-compatible and Multi-AZ resilient object storage service. Ensuring high availability for your data

Object Storage

Cold Storage class to secure long-term object storage. Ideal for deep archived data.

Scaleway Glacier

Flexible and reliable storage for demanding workloads

Block Storage

Storage

Secure your cloud resources with ease on a resilient regional network

Virtual Private Cloud

 A single and secure entrance to your infrastructure

Public Gateway

Improve the performance of your services as you grow

Load Balancer

Buy domain names and manage DNS. Find your favourite extensions at a fair price

Domains and DNS

Centralize and simplify your Scaleway IP address management

IPAM (IP Address Manager)

Expose your HTTP services to the internet with security, reliability, and efficiency by design.

Edge Services

Network

New generation of Relational Databases designed to scale on-demand

Managed Database for PostgreSQL & MySQL

Accelerate your web application with powerful caching of Memory Databases

Managed Database for Redis™

Drive your own document-oriented database. Let us managed the engine

Managed MongoDB®

Go serverless with fully managed database

Serverless SQL Database

Send messages and events without having to manage your message broker

Messaging and Queuing

Speed up data processing over very large volumes of data with an Apache Spark™ managed solution.

Distributed Data Lab

Data

Hosting for individuals, professionals, and everyone in between.

Web Hosting

Ship your applications only in a few clicks.

Web Platform

Instant delivery of your transactional emails

Transactional Email

Monitor infrastructures in minutes with a fully managed observability solution

Cockpit

A purpose-built bridge between connected hardware and cloud.

IoT Hub

Business Applications

The easiest way to safely collaborate in the cloud

Identity and Access Management (IAM)

Protect your sensitive data across your cloud infrastructure

Secret Manager

Easily track your consumption in an all-in-one tool

Cost Manager

Accurately track your environmental impact and make informed choices

Environmental Footprint Calculator

Security & Organization

Scaleway API

Deploy and manage your infrastructure directly from the command line

Securely and efficiently provision and manage Infrastructure as Code with Terraform

Terraform

Developer Tools

Data & Tools

Artificial Intelligence

Public Sector

Gaming

Media and Entertainment

Retail and E-commerce

Startup Program

Financial Services

Industrial

Industries

Cloud Storage Solutions

Kubernetes Solutions

Serverless Applications

Use Cases

Managed Web Hosting

Dedicated Web Hosting

Solutions

All products

Product updates

Betas

Changelog

Blog

Ecosystem

Slack Community

Feature Requests

Scaleway Learning

Community

Events

Marketplace

Careers

About us

Labs

Customer Testimonials

Company

Resources

Pricing

Docs

Contact

_The second edition of ai-PULSE, held at Paris's iconic Station F on November 7, kicked off with a powerful morning session that highlighted some of the most important voices in AI today. The morning’s keynote speakers shared their insights into the latest advancements, the growing power of open-source AI, and the urgent need for Europe to position itself as a hub for sovereign, open, and efficient AI solutions._


## A future shaped by open, efficient, and sovereign AI

Europe is not only catching up with AI powerhouses in the US and China, but is emerging as a formidable force in its own right. **Xavier Niel, Founder of Iliad Group,** began his address with a bold statement: "18 months ago, when we said we wanted to create a tech AI conference in Paris, people told us, 'Guys, you're completely crazy, Europe has no AI ecosystem!' Today, we are proving them wrong." He added, "thanks to our talent, we’re building amazing startups like **Mistral, Poolside,** and **Argil.** And to keep that talent, we’ve created **Kyutai**, where we’re assembling the **‘Avengers of AI’,** bringing together the best minds to develop open source AI models in Europe, built according to our values and regulations."

![Xavier Niel](https://www-uploads.scaleway.com/AI_Pulse_0443_2a6832e2ad.jpg)

One of the most exciting innovations to emerge from **Kyutai** is [Moshi](https://moshi.chat/), an open source conversational AI model. "Moshi is the first truly **open-source AI model** created in Europe, and it’s completely free to use. It’s a game-changer for the industry," **Niel** declared. 

**Damien Lucas, CEO of Scaleway** and **Aude Durand, Deputy CEO of iliad Group,** took to the stage next, declaring that **Moshi** is now available as a one-click deployment service through Scaleway's platform.

**Durand** then emphasized the theme of this year’s event: "Big, Efficient, Open." "In AI, we know that top-tier models require immense computational power," she said. **Lucas** added: ”Scaleway is proud to support **cutting-edge AI models** by offering access to **3000 NVIDIA H100 GPUs today** - triple our total this time last year - and in the coming weeks, **we will surpass 5000 GPUs.** This will allow developers across Europe to leverage the power of AI, while minimizing **environmental impact.**"

![Aude Durand and Damien Lucas](https://www-uploads.scaleway.com/Aude_Durand_and_Damien_Lucas_5e4ed468a5.jpg)

As part of its commitment to sustainable AI, **Scaleway** has also launched its new [Environmental Footprint Calculator](https://www.scaleway.com/en/environmental-footprint-calculator/). "Transparency matters," said **Lucas.** "We want to give developers the tools to understand the environmental impact of their AI workloads, covering everything from emissions to water consumption. It’s the most comprehensive calculator available today, and it’s already in beta." Lucas also unveiled [Generative APIs](https://www.scaleway.com/en/generative-apis/), a new Scaleway service which allows developers to move from Open AI to alternative open source models in just one line of code. More about Scaleway's ai-PULSE announcements [here](https://www.scaleway.com/en/news/scaleway-affirms-ais-big-efficient-and-open-future-at-ai-pulse-2024/).

**Michael Dell, Chairman and CEO of Dell Technologies,** then shared his thoughts on the accelerating pace of AI innovation, drawing comparisons with the rise of the internet. "If you remember **the internet in the mid-90s**, people were wondering what it really meant,” he said. “It took about a decade before the full impact was understood. We're in that same phase with AI today." **Dell** then asserted his belief in the transformative power of AI, both in the cloud and on-device: “AI will be used **both on the cloud and on devices.** It will be used at the edge, on the point of activity closest where the action is. It will run on cloud platforms, on your phone, in cars, factories, and retail stores,” he said. 

**Dell** also spoke passionately about open-source AI, emphasizing that it is a critical driver of innovation. "**Open-source AI has already revolutionized the field**, and it’s just getting started," he enthused. "**Last year would’ve been more about closed models; the world has shifted now. Two thirds of activity is now happening in open models**; derivative, small models that don’t necessarily need to run on massive clusters."

 
![Michael Dell, Dell](https://www-uploads.scaleway.com/AI_Pulse_0578_a46729ee39.jpg)

As AI continues to evolve, **Dell** predicted "we’re transitioning **from calculating to thinking.** The future of AI will require new architectures that bring together data, computing, and cognition in a way that helps humans become more successful," he explained.


## The road to power-efficient AI

**Renee J. James, Chairman and CEO of Ampere Computing,** took the stage next to talk about the future of AI hardware. She reflected on the importance of power efficiency in AI infrastructure. “We set out to build something twice as powerful, but with half the energy consumption, and we ended up getting **3 to 4 times the performance,**” said **James.**

![Renee J. James, Ampere Computing](https://www-uploads.scaleway.com/AI_Pulse_0822_07e72ff1c0.jpg)

In a speech echoing her CPO **Jeff Wittich’s** [assertion last year](https://www.scaleway.com/en/blog/why-cpus-also-make-sense-for-ai-inference/) that Ampere CPUs could handle many inference workloads whilst using 3-5 times less energy than comparable GPUs, she stressed that **AI hardware, particularly processors, need to evolve to meet the increasing demands of AI models.** “Today, we are all about training models, we have people building unlimited clusters to train,” she explained. “But as [AI] starts to moderate and become mainstream, **we have to make it affordable and environmentally efficient,** because we don’t have more power. That's the journey we are going to begin.”

**James** also pointed out that **Europe has an untapped wealth of intellectual talent,** especially in computer architecture, and she believes the continent must focus on increasing investment levels to continue pushing the boundaries of AI hardware. “Access to capital is one of the most important things that needs to continue to improve and the ability for investors to take more risks,” she concluded.

![Renen Hallak, VAST Data, and Aude Durand](https://www-uploads.scaleway.com/AI_Pulse_1195_498a6c0c48.jpg)

**Renen Hallak, Founder and CEO of VAST Data,** agreed with James on the growing importance of efficient data management in **AI scaling.** “AI is changing from allowing us to analyze numbers and rows and columns to pictures and videos, natural language and sound. Older systems were not built for that. So we need to break that paradigm, and that requires a **new software architecture,**” said **Hallak.** He then discussed the shift from CPUs to GPUs for data analysis, emphasizing the need for new software architectures “when you build exabyte-scale clusters that need to manage themselves.” **Hallak** also addressed the growing role of AI agents in problem-solving. “**AI agents will soon be communicating with each other, generating ideas, and solving problems that humans cannot solve on their own,”** he said.

**Bryan Catanzaro, VP of Applied Deep Learning Research Research at NVIDIA,** and **Jean-Baptiste Kempf, Scaleway’s Chief Technical Officer,** shared insights on the future of AI infrastructures. Catanzaro discussed the challenges of scaling AI models, particularly when it comes to GPU clusters. “**We’re going to find that instead of putting a million GPUs in a single data center, we're going to run training and inference in a more distributed way, so we can have better access to electricity, and scale in a more efficient way**,” he said, highlighting the importance of optimizing AI for **both performance and sustainability.**

![Bryan Catanzaro, NVIDIA, and Jean-Baptiste Kempf](https://www-uploads.scaleway.com/AI_Pulse_0947_7cb28e1cf6.jpg)

**Kempf echoed** this sentiment, explaining that new AI models will need to be designed with compute efficiency top of mind. “We talked about thousands or hundreds of thousands of GPUs for training, but **90% of the compute power will actually be dedicated to inference.** So, I don't believe we’ve reached the maximum capacity for inference yet,” he said.

**Catanzaro** added, “we change everything at NVIDIA. My job is to figure out how to change our models and make them much more efficient, both for training and deployment. **NVIDIA is not just a chip company; we focus on accelerating computing through full-stack optimization.** The level of specialization that goes into our GPUs for AI is something we decide on every day, focusing most importantly on machine learning and intelligence.”


## Multimodal foundation mode and groundbreaking voices in AI

**Patrick Perez, CEO of Kyutai,** then spoke about the groundbreaking work behind **Moshi**. “Since launching in mid-September, **Moshi has already been experienced by over half a million people**,”said Perez. He explained that Moshi is more than just a conversational AI; it is a **multimodal foundation model** capable of performing various tasks, from speech synthesis to question answering. “What we’ve created with Moshi is not just a voice model, but a platform for developers to innovate on top of,” he added. **The open-source nature of Moshi is a key part of Kyutai’s mission to democratize AI and ensure that European models can compete on the global stage.**

![Patrick Perez, Kyutai](https://www-uploads.scaleway.com/AI_Pulse_1261_f265aaf8ab.jpg)

**Charles Kantor, CEO of H** then addressed the potential of action models in AI, highlighting the importance of **combining LLMs (Large Language Models) and VLMs (Vision Models) for specialized tasks.** "Action models are incredibly powerful today. By orchestrating LLMs that generate plans and workflows and VLMs that understand interfaces like desktop or Android, you can create systems that can take action effectively," said Kantor. He also emphasized the need for **sector-specific focus**: "the goal is to focus on one industry and master it, building a holistic system with well-orchestrated layers."

![Charles Kantor, H](https://www-uploads.scaleway.com/AI_Pulse_1478_966c47ed21.jpg)

## Building an inclusive AI ecosystem

**Clara Chappaz, France’s Secretary of State for AI and Digital Affairs,** concluded the morning's sessions by discussing the importance of building a distinctive AI ecosystem in Europe. "We are at the very beginning of this technology. **It’s crucial to ensure AI is inclusive, frugal, and ethical**," said Chappaz. She praised **France’s strong educational system for producing top AI talent,** and emphasized the role of infrastructure and private companies, like **Scaleway,** in providing the necessary tools for that talent to thrive and launch companies in France. 

![Clara Chappaz](https://www-uploads.scaleway.com/AI_Pulse_1616_f2ee613181.jpg)

With AI’s growing energy demands, she mentioned the need for green energy in powering infrastructures. “**We [in France] have some of the greenest (lowest carbon) energy in the world.** You all know how much these machines consume when it comes to power and we need to think about how we feed these models with energy that is green, how do we continue fostering these technologies without damaging the planet.” 

![Aude Durand, Clara Chappaz & Damien Lucas](https://www-uploads.scaleway.com/AI_Pulse_1713_e3eb7c2284.jpg)

For Chappaz, what's required is a **'third way' for AI;** a new, collaborative, and inclusive approach to AI development, distinct from the competition-driven dynamics seen in the US and China. “The way we want to build this ecosystem and want to use AI is a way that can resonate with a lot of people. We like to call it the third way; one that brings people together," she concluded. 


- _This is an extract from Saleway's forthcoming **ai-PULSE 2024 white paper**, which will wrap up all of the event's key AI insights. Watch this space!_
- _Meanwhile, you can check out replays of all of the above sessions, plus many more besides, on our [YouTube channel](https://www.youtube.com/channel/UC-Le3MdpQ79hWz8_t2_pGEw). Enjoy!_


large_GenericAnnounce-SphereCrop-ai-PULSE-2024-Illustration-Blog.jpg

medium_GenericAnnounce-SphereCrop-ai-PULSE-2024-Illustration-Blog.jpg

small_GenericAnnounce-SphereCrop-ai-PULSE-2024-Illustration-Blog.jpg

thumbnail_GenericAnnounce-SphereCrop-ai-PULSE-2024-Illustration-Blog.jpg

GenericAnnounce-SphereCrop-ai-PULSE-2024-Illustration-Blog.jpg

best-quotes-ai-pulse-2024-keynotes

Kubernetes has become a game-changing technology for container orchestration, and the vibrant open-source community has been a driving force behind its success. The curated selection of Kubernetes tooling presented here shows the breadth and depth of the ecosystem that has grown around Kubernetes.

From automation and monitoring to networking and security, these tools provide a solid foundation for managing containerized applications. The ongoing development of new open-source projects within the Kubernetes community is a testament to its vitality and innovation.

If you have spend countless hours working with Kubernetes, you understand the unique challenges of managing a complex infrastructure at scale. While Kubernetes is an incredibly powerful platform for deploying and managing containerized applications, it can also transform into a complex mess. So let's discover new tools to orchestrate your infrastructure!

![](https://www-uploads.scaleway.com/Kubernetes_tooling_Scaleway_5cef6c71ac.gif)

## Deployment and configuration management 

### Kustomize
[Kustomize](https://kustomize.io/) is a tool that lets you customize Kubernetes resources by applying patches to base configurations. It provides a simple and declarative way to manage your Kubernetes configurations, allowing you to manage different environments and configurations with ease. 
👉 [Kustomize’s GitHub](https://github.com/kubernetes-sigs/kustomize) 

### Jsonnet
[Jsonnet](https://jsonnet.org) is a configuration language that can be used to generate Kubernetes manifests. It provides a clear way to define Kubernetes resources using a flexible and expressive syntax, which can be used to manage complex and large configurations. 
👉 [Jsonnet's GitHub](https://github.com/google/jsonnet)

### Skaffold
[Skaffold](https://skaffold.dev/) is a command-line tool that streamlines the development workflow for Kubernetes applications. It supports many popular programming languages and development tools, and can be integrated with CI/CD pipelines.
👉 [Skaffold's GitHub](https://github.com/GoogleContainerTools/skaffold)

### Helm
Helm is the well-known package manager for Kubernetes that simplifies the installation and management of complex applications. It provides a templating system that allows you to define your applications as charts, which can be effortlessly installed and configured on your Kubernetes cluster. 
You can also read more on our [blog post about Helm](https://www.scaleway.com/en/blog/helm-guide/).
👉 [Helm's GitHub](https://github.com/helm/helm) 

### Kubeconfig-generator
[Kubeconfig-generator](https://github.com/kubernetes/kubeconfig-generator) is a tool for generating Kubernetes configuration files (kubeconfig) for multiple clusters and users. With Kubeconfig-generator, you can simplify the process of managing multiple clusters and users by generating configuration files automatically. It supports many popular Kubernetes distributions and can be integrated with CI/CD pipelines. 
👉 [Kubeconfig-generator's Github](https://github.com/kubernetes/kubeconfig-generator) 



## Monitoring and observability 

### Prometheus
[Prometheus](https://prometheus.io) is a monitoring tool that collects metrics from Kubernetes resources and stores them in a time-series database—it is a powerful way to monitor the health and performance of your Kubernetes cluster and applications. 
👉 [Prometheus's GitHub](https://github.com/prometheus/prometheus) 

### Thanos
[Thanos](https://thanos.io) is a tool that extends Prometheus by adding long-term storage, global querying, and cross-cluster federation capabilities to store your Kubernetes metrics over a long period of time. 
👉 [Thano's GitHub](https://github.com/thanos-io/thanos) 

### Grafana
[Grafana](https://grafana.com) is a visualization tool that can be used with Prometheus to create dashboards and visualizations of your Kubernetes metrics, making it smoother to understand the health and performance of your applications. 
👉 [Grafana's GitHub](https://github.com/grafana/grafana) 

### Kubevisor
[Kubevisor](https://kubevisor.io) is a lightweight and extensible Kubernetes resource monitor that provides real-time insights into the health and performance infrastructure. You can visualize resource usage and metrics, set alerts and thresholds, and troubleshoot issues quickly. Kubevisor supports many popular Kubernetes distributions and can be integrated with Prometheus and Grafana. 
👉 [Kubevisor's GitHub](https://github.com/kubevisor/kubevisor)


### Sysdig
[Sysdig](https://sysdig.com) unifies visibility, security, and compliance for Kubernetes and other cloud-native technologies. It allows you to monitor your Kubernetes infrastructure and applications in real-time, detect security threats and vulnerabilities, and enforce compliance policies. Sysdig provides a set of features for monitoring and observability, including container visibility, network security, and compliance auditing. 
👉 [Sysdig's GitHub](https://github.com/sysdiglabs) 



## Logging and tracing 

### Elasticsearch
[Elasticsearch](elastic.co) is a search engine that can be used to store and search logs generated by Kubernetes resources. It simplifies the management and search of logs in Kubernetes to troubleshoot issues in your applications. 
👉 [Elasticsearch's GitHub](https://github.com/elastic/elasticsearch)

### Fluentd
[Fluentd](https://www.fluentd.org) is a data collection tool that can collect logs and send them to Elasticsearch or other destinations. 
👉 [Fluentd's GitHub](https://github.com/fluent/fluentd) 


### Jaeger
[Jaeger](https://www.jaegertracing.io) is a distributed tracing system that can be used to monitor the performance of your Kubernetes applications by tracing requests through your Kubernetes cluster, making it easy to identify performance bottlenecks and troubleshoot issues. 
👉 [Jaeger's GitHub](https://github.com/jaegertracing/jaeger) 

### OpenTelemetry
[OpenTelemetry](https://opentelemetry.io) provides a vendor-neutral and language-agnostic way to collect, process, and export telemetry data to collect logs, traces, and metrics from your  applications and infrastructure, and export them to popular observability platforms such as Prometheus and Grafana. 👉 [OpenTelemetry's GitHub](https://github.com/open-telemetry) 

### Fluent Bit
[Fluent Bit](https://fluentbit.io) is a lightweight and high-performance log and event collector that supports many popular logging systems and formats. With Fluent Bit, you can collect and process logs from your Kubernetes applications and infrastructure, and export them to popular logging platforms such as Elasticsearch and Splunk. 
👉 [Fluent Bit's GitHub](https://github.com/fluent/fluent-bit) 


## Networking and service discovery 

### CoreDNS
[CoreDNS](https://coredns.io) is a DNS server that can be used to provide service discovery within your Kubernetes cluster to manage DNS records for your Kubernetes resources, making it easy to discover and communicate with services within your cluster. 
👉 [CoreDNS's GitHub](https://github.com/coredns/coredns)

### Traefik
[Traefik](https://traefik.io) is a reverse proxy and load balancer that can be used to manage network traffic between Kubernetes resources to manage network traffic and enforce security policies to secure and optimize communication between your Kubernetes resources. 
👉 [Traefik's GitHub](https://github.com/traefik/traefik)

### Istio
[Istio](https://istio.io) is a service mesh that can be used to manage and secure communication between services within your Kubernetes cluster to manage traffic routing, load balancing, and security policies. 
👉 [Istio's GitHub](https://github.com/istio/istio)

### Contour
[Contour](https://projectcontour.io) is a Kubernetes ingress controller that provides a high-performance and scalable way to manage incoming traffic. Contour supports many popular features, such as TLS termination, load balancing, and path-based routing. 
👉 [Contour's GitHub](https://github.com/projectcontour/contour)

## Security 

### Kubernetes Secrets
[Kubernetes Secrets](https://kubernetes.io/docs/concepts/configuration/secret) is a tool that can be used to store and manage sensitive information like passwords and API keys. It provides a simple and secure way to manage secrets within your Kubernetes cluster to protect sensitive information. 
👉 [Kubernetes Secrets's GitHub](https://github.com/kubernetes/kubernetes) 

### RBAC (Role-Based Access Control)
[RBAC](https://kubernetes.io/docs/reference/access-authn-authz/rbac) is a security mechanism that can be used to control access to Kubernetes resources based on roles and permissions to manage access control within your Kubernetes cluster, making it easy to enforce security policies. 
👉 [RBAC's GitHub](https://github.com/kubernetes/kubernetes) 

### Open Policy Agent (OPA)
[OPA](https://www.openpolicyagent.org) is a policy engine that can be used to enforce security policies and access control within your Kubernetes cluster to define policies, which can be used to enforce security and compliance requirements. 
👉 [OPA's GitHub](https://github.com/open-policy-agent/opa) 

### Kyverno
[Kyverno](https://kyverno.io) is a policy engine designed for Kubernetes to ensure compliance and security for your Kubernetes infrastructure and applications. Kyverno is easy to deploy and configure and provides a powerful set of features for policy enforcement. 
👉 [Kyverno's GitHub](https://github.com/kyverno/kyverno)

### Gatekeeper
Gatekeeper is a policy engine designed for Kubernetes. It provides a way to enforce policies and rules for Kubernetes resources using the Open Policy Agent (OPA) engine. With Gatekeeper, you can ensure compliance and security for your Kubernetes infrastructure and applications. 
👉 [Gatekeeper's GitHub](https://github.com/open-policy-agent/gatekeeper)


## Continuous integration/Continuous deployment 

### Dagger
[Dagger](https://dagger.io/) is a Kubernetes-native deployment tool that simplifies the deployment and management of applications on Kubernetes. With Dagger, you can define and deploy complex applications in minutes, without the need for complex YAML files or custom scripts. It provides a simple and intuitive user interface for managing deployments, with real-time monitoring and logging. 
👉 [Dagger's GitHub](https://github.com/dagger/dagger)


### Jenkins
[Jenkins](https://www.jenkins.io) is a popular CI/CD tool that can be used to automate the process of building, testing, and deploying your applications to Kubernetes. It provides a simple and flexible way to define pipelines, which can be used to automate the entire software delivery process. 
👉 [Jenkins's GitHub](https://github.com/jenkinsci/jenkins) 

### GitLab
[GitLab](https://about.gitlab.com) is a platform that provides CI/CD pipelines for building, testing, and deploying your applications to Kubernetes. It provides a simple and powerful way to manage your source code and automate the software delivery process.
👉 [GitLab's GitHub](https://github.com/gitlabhq/gitlabhq) 

### Argo CD
[Argo CD](https://argoproj.github.io/cd/) is a tool that can be used to automate the deployment of your applications to Kubernetes by using GitOps principles. It provides a simple and powerful way to manage the deployment of your applications, making it easy to ensure consistency and reliability. 
👉 [Argo CD's GitHub](https://github.com/argoproj/argo-cd) 


### Tekton
[Tekton](https://tekton.dev) is an open source framework for building CI/CD pipelines on Kubernetes to define and execute pipelines using Kubernetes resources. Tekton is gaining popularity as more organizations adopt Kubernetes for CI/CD. 
👉 [Tekton's GitHub](https://github.com/tektoncd/pipeline)


## Infrastructure-as-Code (IaC)

### Terraform
[Terraform](https://www.terraform.io) is a tool that can be used to define and manage your Kubernetes infrastructure as code, making it easy to provision and manage your Kubernetes cluster and associated resources.
Learn how to deploy your Terraform module in a few click with our [guided hands-on here](https://www.scaleway.com/en/blog/terraform-module-deploy-your-infrastructure-in-one-click/).
👉 [Terraform's GitHub](https://github.com/hashicorp/terraform) 

### Ansible
[Ansible](https://www.ansible.com) is a tool that can be used to automate the provisioning and configuration of your Kubernetes resources. It provides a simple and powerful way to manage your infrastructure, making deploying and managing your Kubernetes resources easy. 
👉 [Ansible's GitHub](https://github.com/ansible/ansible) 

### Pulumi
[Pulumi](https://www.pulumi.com) is a tool that can be used to create and manage your Kubernetes resources using familiar programming languages like Python and JavaScript. 
👉 [Pulumi's GitHub](https://github.com/pulumi/pulumi)

### Crossplane
[Crossplane](https://crossplane.io) is an open source platform for managing cloud-native infrastructure that helps to manage infrastructure across multiple clouds. Crossplane is gaining popularity as more organizations [adopt multi-cloud strategies](https://www.scaleway.com/en/blog/k8s-multi-cloud/). 
👉 [Crossplane's GitHub](https://github.com/crossplaneio/crossplane) 


## Final thoughts

The momentum Kubernetes has right now is driving so many positive things. Every vendor is taking a Kubernetes-first approach to building their solutions and tools. For example, modern applications are being built and packaged to run on Kubernetes, and most innovations in areas like security and development are being built around Kubernetes. As well as this, every major cloud offers its own Kubernetes services. The ecosystem is huge, it’s still growing, and it’s very welcoming. That last bit is helping lower the entry barrier and allowing more people to enter.

As Kubernetes continues to evolve, we can expect even more exciting developments and advancements to emerge. It's an exciting time to be part of the Kubernetes community, and we look forward to seeing the new tools and solutions that will be developed in the future.










large_Kubernetes blog1.png

medium_Kubernetes blog1.png

small_Kubernetes blog1.png

thumbnail_Kubernetes blog1.webp

Kubernetes blog1.webp

kubernetes-tooling-open-source

Today marks one year since Scaleway’s Serverless Functions and Containers entered General Availability. We’ve achieved a lot in the past year, and we’re only just getting started. In the serverless team, we don't see serverless as a single product, but as an ecosystem, driven by a philosophy of making the cloud easier to use, more efficient, but also more affordable to use., we are committed to delivering a set of services and frameworks that provide a complete serverless experience to a diverse range of applications. In this post we’ll describe the big ideas behind the serverless ecosystem, where we are today, and where we’re going in the future. 

## Making the cloud feel like one big operating system 
Writing and running code on a laptop is exactly that, writing the code, then telling the operating system to run it. Developers don't need to specify how to configure or share the underlying resources, which bits of the system to interface with, or what limits they want to enforce. It’s so simple that even children can do it. Serverless is about bringing this same simplicity to the cloud, making it as simple as running an application on your laptop. This means users can focus on writing code, while the serverless platform handles the rest. This means provisioning, configuring, and scaling whatever underlying infrastructure is needed, billing the user only for what they use, and cleaning up afterwards. 

We still have a way to go before we achieve the serverless dream, and today serverless is still synonymous with just two products: function-as-a-service, and container-as-a-service. Both represent a significant step forward in shielding the user from complexity, allowing them to focus on writing code, and billing them only for what they use. However, functions and containers are just the foundations of the larger serverless ecosystem. To provide a complete serverless experience, we must also consider how we compose functions and containers into larger workflows, how we manage state and scale our storage, how we can make it easy to port existing applications, and how we can write new applications for this novel environment. 

## How we built our Serverless foundations: Functions and Containers 

Just as you can’t build a house on shaky foundations, you can’t build a serverless ecosystem without a solid base. In serverless, this base is Functions-as-a-Service and Containers-as-a-Service. These are the workhorses of serverless, executing user’s code, and scaling on demand. On top of this base of FaaS and CaaS, we can build the rest of our serverless ecosystem. We started work on our FaaS and CaaS products as early as 2019. 

Although we were quick to get to an MVP, GA was still almost two years away. These two years were spent on building the scaffolding around the core compute product, to create a secure, smooth, production-grade serverless product. Significant projects included the build-and-deploy pipeline, configuring and stress-testing our autoscaling, performance tuning and profiling, scaling the platform across multiple clusters in multiple regions, security audits, and integrating the product into the Scaleway ecosystem. At the time, the company was focused on bare metal and virtual machines, so setting up and running a scalable serverless infrastructure was a big change in mindset. 

It was also a big technical change, as we were sharing infrastructure between tenants at a much higher granularity, and dealing with high-volume, short-lived workloads. This placed increased stress on our metrics, logs, and internal APIs, and required a rethink of how we do observability and billing. It also put a strain on our nascent managed Kubernetes Service, with the volume of containers regularly causing headaches for the Container Registry Service. We launched the working beta in the summer of 2020. 

The platform was slow during the first weeks due to unexpected consumption from the beta testers and we ran into several scaling issues. Through more than a year of beta tests, we gathered lots of feedback on how we could build an even better and more innovative product. Some of the ideas that came out of this process are still in the pipeline, so watch this space! Finally, after two years of work and a last minute bug that almost crashed the product, we were happy to release our serverless platform to the world. Scaleway’s Serverless Functions and Containers entered GA in November 2021. 

Since then, we have continuously added features to make our products more performant, flexible, powerful, and easy-to-use. Although we will continue to add features to make these foundations even better (such as support for public and private container registries, new language runtimes, and improved logging and metrics), we are now in the next phase of building our serverless ecosystem. 

## What’s next in the serverless ecosystem? 

Now that we have a solid foundation in place with Serverless Functions and Containers, we can look to the future of serverless, and what that means for the rest of the ecosystem at Scaleway. We have worked closely with existing users of both serverless and non-serverless products to determine how our serverless ecosystem can best suit their needs. As a result of these consultations, we are now working on a number of products and enhancements that will allow users to build more complex applications, at larger scale, with more data, and more resources. We can break this work down into 5 categories: workflows, storage, programming, developer experience, and performance. ## Workflows Although users can achieve a lot with single functions and containers, the real power of serverless comes from orchestrating multiple functions and containers as part of larger workflows. These workflows can be used for any multi-stage task, from downstream data processing pipelines, to multi-stage image compression and analysis, to distributed machine learning training. Serverless workflows cannot be configured without suitable plumbing to pass data between stages. This is where Scaleway’s new Messaging and Queueing service comes into play. Part of the next phase of our serverless ecosystem is a serverless orchestrator, built on messaging and queueing, which lets users connect functions and containers to a number of different event sources, including queues, publish-subscribe, and Amazon S3 events. 

## State and storage 
Compute isn’t much use without access to data, so in addition to serverless functions and containers, we need to provide serverless storage. Serverless storage shares many of the properties as serverless compute, namely users don’t have to provision or scale the underlying infrastructure, and they pay only for what they use. Object storage has provided a form of serverless storage for several years at Scaleway, but object storage is only part of the story. To broaden our serverless storage offering at Scaleway, we are working on a new line of serverless NoSQL/SQL databases, which will offer the same auto-scaling, pay-as-you-go experience as Serverless Functions and Containers. 
With combined serverless compute and storage, users can build front-to-back data driven applications without configuring any infrastructure. For example, Lego only uses serverless resources (functions, databases, storage) and managed services (API gateway, Messaging, Emailing …) to run its whole e-commerce platform. This enabled them to create an efficient applications that use few resources most of the year while being able to scale to match customers demand during the Black Friday or the Christmas season. 

## Building new applications and porting existing applications 

Today’s serverless programming model is built on stateless functions. Each function can’t communicate with others, and must be able to operate at arbitrary scale. While this is a powerful tool for certain applications, it makes it difficult to port many existing applications, and presents a steep learning curve for non-serverless developers. Indeed as of today, to maximize Serverless benefits, one must think in terms of small single-purpose functions and how to orchestrate them. As is the case in other distributed and parallel execution environments, we want to support a range of different programming models and applications, not just those built on stateless functions. 
To do this, we are focusing on two themes: a) building new programming models specifically for serverless b) supporting existing applications written for other execution environments. New serverless programming models are those that take advantage of the effectively limitless scaling of the serverless execution environment, inspired by ideas from actor-based parallelism, and messaging passing frameworks. These often cater to distributed, highly parallel applications such as scientific simulations and distributed ML training built on MPI. Porting existing applications means providing seamless integration with other parallel and distributed programming frameworks. Applications built on top of these frameworks can then change a line of config, to switch from their existing execution environment, to the new serverless backend. Two good examples of such frameworks are Spark and Dask, used extensively by data scientists. By porting these frameworks to serverless, we can open up the potential of serverless to a large number of new users and use-cases. 

## Developer experience 
The serverless developer experience is different from that encountered in many other execution environments. By definition, serverless applications are distributed, and run in an environment that is controlled by the provider. This exacerbates challenges around logging, debugging, and monitoring, similar to those found in microservice architectures, especially those relying on Cloud Service Providers services. In particular to verify that the service (function) developed still works in the provider's environment users are required to deploy it, increasing the time to get debugging information hence breaking the code/run/improve loop. 

To improve the serverless developer experience and fulfill the promise of a seamless experience, we are working on a number of tools and enhancements. The first of these is offline development and testing. The serverless execution environment is defined and controlled by the provider, and so replicating it on a local development machine can be challenging. 
We are building tools to help developers get meaningful feedback from their development environment, without having to redeploy functions remotely. The second improvement we are making in this area is monitoring and logging. By integrating with Scaleway’s Observability Platform-as-a-service, we will provide users with a detailed overview of their serverless functions, along with logs and metrics that can help them manage this distributed environment. 

## Performance 
Performance in serverless is often distilled down to a single problem, that of cold starts. Although cold starts are an important factor, especially in user-facing applications, serverless performance also covers scale-out latency, build times, and latency when interacting with the rest of the ecosystem. These performance concerns are multiplied for every container and function that is added to an application, so it’s important for us to ensure high performance when considering serverless at scale. Our work in this area covers two areas: reducing the build, deployment and invocation time of our existing architecture, and researching new architectures, such as those based on WebAssembly. Reducing the latency in our existing architecture involves caching, pooling and simply reducing the resources needed to execute each function. By adding registry caches, reusing isolation environments between requests, and pooling pre-warmed VMs and containers, we can reduce the latency involved in all aspects of our serverless systems. Experimenting with lightweight isolation mechanisms such as WebAssembly, offers orders of magnitude improvements in cold start times, as well as exciting opportunities to build new custom runtime environments, with low-latency messaging, and shared memory between functions. Our research and development on new serverless runtimes is ongoing, but we hope to reveal some exciting new improvements next year. 

## Final thoughts 
Serverless is a philosophy for building cloud platforms that feel like one big operating system. At Scaleway we’re committed to delivering an ecosystem of products that adhere to this philosophy, allowing users to build large, scalable, stateful applications as simply and affordably as possible. In one short year we’ve built two products that form the foundation of this ecosystem: Serverless Functions and Serverless Containers. There’s much, much more to come, so watch this space, and thank you for joining us on the journey.

large_Corpo-Space-Illustration-Blog.webp

medium_Corpo-Space-Illustration-Blog.webp

small_Corpo-Space-Illustration-Blog.webp

thumbnail_Corpo-Space-Illustration-Blog.webp

Corpo-Space-Illustration-Blog.webp

1-year-anniversary-serverless-blog-post

_As CPO of US-based chipmaker [Ampere Computing](https://amperecomputing.com/), Jeff Wittich has an important message for IT executives: artificial intelligence inference doesn’t necessarily need supercomputers, or GPUs. In many cases, he claims, CPUs are not only good enough, they’re even ideal. Why? Because they can offer right-sized compute power with minimal energy consumption, thereby limiting AI’s impact on the planet and on cloud budgets. We spoke to Wittich ahead of his keynote at [ai-PULSE](https://www.ai-pulse.eu/) on November 17…_

## How does Ampere want to be considered by cloud providers today when it comes to AI? 
**Jeff Wittich**: Ampere’s mission from day one has been to deliver sustainable computing for modern performance environments like the cloud. That extends to AI too. Cloud service providers (CSPs) should consider Ampere for all needs in the cloud, including when looking to build AI workload capabilities. 

We know one of CSPs’ biggest challenges is power consumption. Using more power is costly, plus power is scarce, and you can’t expand your data center infinitely. This means **we need to deliver more efficient systems over time, to provide more compute capacity without consuming more power**.

AI inference has really brought this into the forefront, as demand for it has increased rapidly, making that power challenge even more difficult to solve. We have a solution that tackles that.

Often when we talk about AI, we forget that AI training and inference are two different tasks. 

**Training** [or teaching the AI model with large quantities of data] is a one-off, gigantic task that takes a long time; and for that one time, you might be OK to use the considerable amounts of power required by GPUs and supercomputers.

**Inference** [or using the trained AI model on a regular basis] is different, as it can be millions of tasks running every second. Inference is your “scale” model, that you’re running all the time, so efficiency is more important here.

So whereas accelerators can make a lot of sense for training, **building inference workload doesn’t need to be done on supercomputing hardware**.

In fact, general-purpose CPUs are good at inference, and they always have been. Our CPUs are especially well-suited to the task because they are high-performance and balanced. Plus you need predictable latency in these cases, and to keep processing close to the core, not have it bouncing around all over the place. Having a lot of cores is useful too, as is flexibility. It may be that AI inference isn’t 100% of what you’re asking a CPU to do. If it can do other things at the same time, you get higher overall utilization.

## How can CPUs be enough for inference, when the current trend is “throwing more expensive, power-hungry, and narrowly specialized hardware at AI”*?
**JW**: AI needs today cover a whole spectrum. What are your project’s compute requirements? Do you need to be inferencing all the time? What about memory bandwidth? **For the vast majority of that spectrum, CPUs will be the right-sized solution**. Some inference needs may have a particularly high memory footprint, and therefore need a GPU.

But I think we’ll see a shift in time to smaller, more versatile solutions. It’s like I could have come to work in a Ferrari today, when what I actually need is a more economical electric vehicle that’ll get me here in the same time.

We’re still in the **hype and research phase for AI**, due to the euphoria around these massive large language models (LLMs), where the instinct is to throw the most possible power at a problem and see what happens. But at some point, **these use cases will mature, and efficiency and sustainability will be the victor**.

Not everyone will be able to pay for a solution like ChatGPT, which features all of human knowledge. We’ll see more specialization of models, as well as refinement of existing models. Overall, models will become smaller, and more focused on specific tasks.

_*A quote from Ampere's recent [white paper](https://info.amperecomputing.com/AI-Whitepaper)._

## What are the most interesting inference use cases for Ampere chips today?
**JW**: We’re already seeing some great examples, from real-time voice-to-text translation in any language, which makes things easier for meetings with colleagues in other countries, or increasing accessibility for hearing-deficient people; or generative AI use cases, like artwork, videos, or simplifying everyday routine tasks. These cases all work well with our CPUs.

More specifically, [Matoha](https://matoha.com) uses Ampere CPUs to power its near-infra-red spectroscopy. This allows them to scan a 30-year-old landfill for waste noone back then thought of recycling. They can scan a bottle, figure out what type of plastic it is, and send it to the right recycling location. And it works with other materials too, like fabrics.

We also have [Red Bull Racing](https://www.redbullracing.com/int-en), the highly successful Formula One team, which uses our processors for pre- and in-race day analysis, to optimize their racing strategies. They have a limited amount of time to run these analyses, using complex models based on past race data. Our CPUs allow them to process a lot of data in a very short time, so they can change strategies in real-time, for example, if the weather changes.

## How exactly do Ampere CPUs transfer training data from Nvidia GPUs, for inference?
**JW**: **It’s a common misperception that you need to run training and inference on the same models**. It’s actually very easy to take one framework and run it on another piece of hardware. It’s particularly easy when you use [AI frameworks like] **PyTorch** and **Tensorflow**; the models are extremely portable. 

We have a whole AI team at Ampere, which has developed software called **AI-O**, that allows us to have compatibility across all AI frameworks. So **there’s no need to adapt data models at all**. Just take a model trained with any GPU, put it on an Ampere CPU and it’ll run great. AI-O does some optimization on the data and processing sides, but you don’t need to use it unless you really want to improve performance. Otherwise, no need for quantization or anything like that. People think (transferring from training GPUs to inference CPUs) is incredibly complicated, but it’s not!

## Can data models be adapted to get maximum performance from Ampere CPUs?
**JW**: Yes, just use the software library we have (AI-O): it’s sophisticated, it gets better results, and it makes sure the way the code is compiled is well-suited to our processors. You’ll get several times higher performance for some models, should you choose that option (but you don’t have to).

Sometimes, there’s an advantage to running at lower precision. So instead of running an FP32 [data model], run the model in something like int8. Our processors support FP32, FP16, Bfloat16, int8… any numerical format you’ll want to run in. In the case of int8, you’re essentially getting four times more performance capacity than FP32, and in many cases you’re not losing any accuracy as a result if doing so. And that’s just as easy to do on our processors as it would be on an Nvidia GPU, or Intel or AMD CPU. 

To make things even easier, we ensure you get full support from our AI engineers. That doesn’t exist with all the manufacturers today: they’ll have hardware support, but not software. Better still: we haven’t had many help requests yet, so we like to think that means our solution just works. We do know **a lot of people are using AI-O: we’ve seen a sevenfold usage increase in the past six months**, so that’s fantastic.

## AI consumes considerable amounts of energy and (indirectly) water. Can you quantify the energy savings of Ampere CPUs vs other GPUs for AI inference? 
**JW**: If you run [OpenAI’s generative speech recognition model] **Whisper** on our **128-core Altra CPU versus Nvidia’s A10 card, we consume 3.6 times less power per inference**. Or for something lower-power, like Nvidia Tesla T4 cards, we consume 5.6 times less.

You also have the cooling aspect: the power you’re drawing turns into dissipated heat. So **doing this with 3.6 less times power means it’s that much easier to cool**. So our hardware doesn’t require super-exotic cooling systems, just standard fans.

Water requirements are harder to calculate because there are so many different ways of cooling data centers. But it’s a fact that the easier a CPU is to cool, the less water you need to cool it. 

## How can Ampere help cloud providers to become sustainability leaders?
**JW**: That’s absolutely our mission, as sustainability is one of CSPs’ main pillars. Most people only see the cost, so if we can provide a more efficient processor, great. But we’re seeing more and more CSPs stepping up and providing sustainability messages too, with energy figures, and carbon consumption of Ampere versus Intel and other chipmakers, and so on. **We encourage CSPs to be vocal about that**. 

We’re tackling how to reduce the amount of energy consumed without asking people to use less compute power. So we should be at the forefront of finding ways to create as little impact as possible. Especially **with AI: you hear about some AI usages causing data centers to double their energy consumption**. We need to pick the right solutions to make sure that doesn’t happen.

## What can we expect from Ampere in terms of future developments in CPU technology, particularly in the context of AI and emerging technologies?
**JW**: Over the next few years, we’ll continue to release CPUs that are more efficient, and deliver ever-higher core counts, as that gives you more and more throughputs for things like AI inferencing. So you’ll see us looking to increase output compute without requiring more incremental power, by adding more cores and increasing memory bandwidth and I/O bandwidth, so that’s perfect for AI inferencing too.

In AI, as we have a team of dedicated engineers, you’ll see us put more new features into our CPUs: **we’ve got some interesting ideas in the pipeline to increase inference performance disproportionately**. The pace of innovation in the AI space is extraordinarily fast. We’re releasing new products extremely quickly for that reason. We’re also learning from how our clients are using our CPUs in AI today, to anticipate innovations we’ll work into products we’ll release very soon. **If you take five years to make this tech, you’re already obsolete**. So this is why we’ve adapted our development cycle.

## What are you most excited about today?
**JW**: Sustainability has to be one. Doing something that has a huge impact globally is really exciting. The cloud has a big emissions footprint, globally speaking, so it’s important we take the lead here, including with regards other industries.

More broadly speaking, I’m excited that we're building a new type of general-purpose compute for the world, which isn’t constrained by the limits of data centers to date. By thinking “What does the cloud need?” we’ve done some really cool things, and that’s why we can deliver such great performance across all CSPs. We have limitless capacity to innovate within our CPUs. **It’s a new generation for the cloud era!**

<br>
_Jeff Wittich presents "The Key to AI's Power Efficiency Revolution" (17:25) at ai-PULSE November 17, followed by a panel with Gladia and Powder, "How to make Inference as cost-efficient, sustainable and performant as possible?", from 17:45. [More info here](https://www.ai-pulse.eu/agenda)..._

Scaleway Blog

Latest articles

Why CPUs also make sense for AI inference - interview with Ampere Computing's Jeff Wittich

Update: Details of fr-par VPC incident & response

Symbolic AI is dead… long live symbolic AI!

How to Optimize LLM Performance with NVIDIA H100 GPUs from Scaleway, by Golem.ai

“As a startup, if your cloud costs too much to run, it can kill the business” - GreenoSoil migration interview

“The way Scaleway develops its products is more ‘cloud-like’, so more suited to our needs” - Henrik Tudborg, KeepFocus

10 best practices to configure your VPC

How to optimize Object Storage security and reliability

2 reasons to upgrade your Kubernetes right now!

How Kubernetes powers Le Fourgon’s e-commerce & delivery reinvention

How can engineers make IT more sustainable? Part 4: What to do next

Scalers go Open Source: “You don’t know how much you bring to the table.”

Most popular articles

The Best quotes from ai-PULSE 2024's keynotes

Top open-source Kubernetes management tools

Serverless: 1 year in, and we’re just getting started