What do you think about the word of the week "serverless" ?
Virtualization and containers in a private or public cloud are needed in companies by providing greater flexibility in implementation and operation, ensuring greater availability and scalability. However, apart from the choice of specialized and turnkey SaaS services, the use of private or public cloud solutions requires a first estimation of sizing (RAM, storage, memory), a fine management of the capacity planning and forces to reserve powers of infrastructure to absorb sudden increases in loads, guaranteeing the success of the application.
In response to these issues, a solution has emerged and made the buzz since 2015: the serverless on public clouds Amazon and Google.
The serverless is as its name suggests to run its application on a server-less infrastructure. In fact, this is obviously not the case. The servers exist but their operation and administration are simply transparent because the cloud provider.
The serverless architecture relies on the services (Backend as a Service) or the functions (Functions as a Service) triggered in an event-driven and ephemeral way and adapted to the execution of back-office batch processes and front-end applications. Local state conversation and strong SLA constraint.
Finally, the last decision engine is based on the desire to decline its IS architecture with the serverless micro-service component. The freedom it brings in terms of specific development must be challenged in a context where business solutions can meet the same expectations.
The serverless architecture, like artificial intelligence or virtual reality, is one of the major trends identified by the Gartner in 2017. It will impact the CIOs, change the practices of the devs and ops as well as improve the delivery, the serverless will be this revolution that will not be missed and this as from 2018.
From TechTarget, to continue your reading, see:
Know when to implement serverless vs. containersBy Tom Nolle 27 Dec 2017
Containers save resources and reduce deploy times. So does serverless. Containers are a great host for microservices. So is serverless. But the deployment options are more different than they are alike.Read More
Serverless, containers shape future of cloud computingBy George Lawton 15 Jan 2018
Experts seem to agree that serverless platforms, containers and the eventual merging of the two will be the biggest cloud trends to follow this year.Read More
Serverless development as a route to freedom from cloud lock-inBy Adrian Bridgwater 03 Jan 2018
Named after the South American waterfall, but with the ‘creative’ use of a lower case first letter, iguazio is a data platform company specialising in continuous analytics and event ...Read More
Prepare for serverless technology with these five tipsBy Kathleen Casey 30 Oct 2017
Serverless computing can boost flexibility and reduce overall cloud costs. To tap into these benefits, though, be sure to carefully design and manage your serverless apps.Read More
Getting to know serverless computing architectureBy Eamon McCarthy Earls 27 Oct 2017
Bloggers explore the dynamics of serverless computing architecture, efficacy and efficiency with endpoint security and use cases for bidirectional forwarding detectionRead More
Choose the right APIs and apps for serverless infrastructureBy Zachary Flower 21 Dec 2017While serverless platforms can reduce both infrastructure complexity and cost, they might not be your best bet for monolithic or multi-cloud apps.Read More
En Fran?ais
La virtualisation et les conteneurs dans un cloud privé ou public s’imposent dans les entreprise en apportant une meilleure flexibilité de mise en ?uvre et d’exploitation, l’assurance d’une meilleure disponibilité et scalabilité.
Cependant, hormis le choix de services SaaS spécialisés et clé en main, l’utilisation des solutions cloud privées ou publiques nécessite une première estimation de dimensionnement (RAM, stockage, mémoire), une gestion fine du capacity planning et oblige à réserver des puissances d’infrastructure pour absorber les brusques montées en charges, gage du succès de l’application.
En réponse à ces problématiques, une solution émerge et fait le buzz depuis 2015: le serverless sur les clouds publics Amazon et Google.
Le serverless consiste comme son nom l’indique à faire tourner son applicatif sur une infrastructure sans serveur. Dans les faits, ce n’est bien évidemment pas le cas. Les serveurs existent mais leur exploitation ainsi que leur administration sont simplement transparentes car à la charge du fournisseur cloud.
L’architecture serverless s’appuie sur les services (Backend as a Service) ou les fonctions (Fonctions as a Service) déclenchés de manière événementielle et éphémère et adaptée à pour l’exécution de traitements batch back-office et d’applications front sans conversation d’état local et de contrainte de SLA forte.
Enfin le dernier moteur de décision se repose sur la volonté de décliner son architecture SI avec la composante micro-service unitaire propre au serverless. La liberté qu’il apporte en termes de développement spécifique doit être challengée dans un contexte où des solutions métier peuvent répondre aux mêmes attentes.
L’architecture serverless, à l’instar de l’intelligence artificielle ou de la réalité virtuelle, est une des grandes tendances identifiées par le Gartner en 2017. Tant elle va impacter les DSI, modifier les pratiques des devs et des ops ainsi qu’améliorer le delivery, le serverless sera cette révolution qu’il ne faudra pas rater et ce dès 2018.
Friendly
Jean-Albert EUDE Thought Leader