Kubernetes uses etcd as it’s main backend database. It stores all configuration data in key-value format in etcd. If you lose etcd, your Kubernetes will stop working. Kubernetes uses etcd to store its actual state and desired state of the system. It then uses etcd watch functionality to monitor changes and based on that reconcile actual v/s desired state.
And it’s not just Kubernetes, there are other adopters like Rook, CoreDNS etc.
Most Important Data Model Design Policy
One of the most important design characteristics of etcd is — nothing gets deleted unless the admin truncates data. By design etcd stores data as multi-version key-value pair…
We are all seeing rise of microservices. While Docker is helping teams with development, Kubernetes has gradually become platform of choice for runtime orchestration of the containers. Many companies have successfully improved use of Kubernetes moving their applications into microservices world and while doing so has effectively increased footprint of the microservices manifold. These organizations now face runtime challenges of efficiently routing, monitoring and securing the networking between these microservices — not because we don’t know how, but because the sheer increase in number and volume calls for a dedicated and efficient solution. …
If you or your team is working with containers and pods, chances are you have heard of Kubernetes. Kubernetes is not new, however for many of us, there’s still some mystery to how it all comes together. Getting the basic concepts right is important to understand it overall which will help you not only in your daily work but also in becoming an expert in this field if you want to.
Here we are going to look at the top 5 concepts one absolutely need to know about Kubernetes:
- What is it
- What kind of workload can we run on…
With the rapid adoption of Microservice architecture pattern for distributed systems, we are seeing how a single monolithic application can get decomposed into multiple loosely coupled services. These services run as docker containers, many times horizontally scaled to multiple instances in production environment and orchestrated by complex and advanced platform engines like Kubernetes.
It’s not always an easy task. While developing and deploying is one part of the equation, another important aspect is managing such huge number of micro services and making sure everything is running fine. At any moment in time, operations and support team need to know what…
Data is important. Data gives birth to Information. Someone somewhere is always making sure our data is stored and secured properly. But it’s not magic. There are loads of products, tools and services that keep the data world alive and thriving. SQL Server is one of those venerable relational database management system — and it’s not just a database anymore.
Let’s find out the latest top features that has enhanced SQL Server for good.
Docker Containers
We can now run SQL Server as docker containers. This is good news for cloud adoption. We have both Windows and Linux (Ubuntu, RHEL etc.) …
We are going to see how to build and run a Kotlin with Ktor Docker image. If you are new to Kotlin or Ktor or both, you can start with this to see how it works. If you are not new to Kotlin, you can see how we can build a light-weight Docker image for Kotlin using the multi-stage builds feature.
Kotlin
https://github.com/JetBrains/kotlin
It is an open-source, statically typed programming language. One of the most famous benefits of Kotlin is it’s multi-platform programming capabilities which reduces time and effort to maintain same code for different platforms like Linux, Windows, Android, iOS…
Many companies are aggressively adopting cloud computing. Out of many cloud computing options, many developers and teams are using serverless. I think many people already know what serverless is, but you can glance through What is Serverless if you want a quick refresher or context.
All the big cloud providers provide platform and solutions to run serverless workload. Each of them provides countless tools, integrations etc. and then you pay as you use. But what if you don’t want to go public just to run some of your code which is part of an overall big application. Or what if…
FastAPI and Uvicorn often comes together for many modern applications and micro services. Running them as Docker containers makes total sense for micro services on any cloud platform.
FastAPI
FastAPI is a well-known Python framework for building APIs based on standard type hints.
Starlette is used for the web parts.
Starlette is a lightweight ASGI framework/toolkit, which is ideal for building high performance asyncio services.
ASGI (Asynchronous Server Gateway Interface) is a successor to WSGI, intended to provide a standard interface between async-capable Python web servers, frameworks, and applications. Where…
Docker is good, but it has its challenges. Two of the most popular concerns traditionally have been managing image size and reducing vulnerability and attack surface of the containers.
It has been a traditional practice to keep separate docker files for each environment or platform for an app e.g. a dev/sdk docker image for developers, a debug or performance test oriented image for qa/test environments and a slimmed down just what you need image for production. There may be more, varies from team to team.
But maintaining multiple docker files is a pain, it’s time consuming and error prone. …
The concept of robot or droid is nothing new. But, most of the times when we think of robots, the most prominent images that flash in our mind are probably what we have seen in the famous movies. Times are changing. Robots are coming to production. CES 2021 show cased some incredible innovations and is clearly showing huge confidence in a world of robotics gradually opening not just in industrial sectors but also in our day-to-day life.
Home Robots
AI is no longer complex algorithms, machine learning and neural networks. AI is now personal and predictive. AI is now in daily life.
…
