[cs615asa] [CS615] meetup Zehui Xie

[Zehui Xie]

Meetup-CS615-zehui xie-zxie11

I joined this Clod compute-related meeting "Running a Cloud-Native Data Tier with K8ssandra - Kubecon Europe 2021!" start on 04/05/2021. This meeting talked about a new tool(K8ssandra) to help us manage data on a k8s based distributed system.

About this Event

I attend this event at "Running a Cloud-Native Data Tier with K8ssandra - Kubecon Europe 2021!" on Tue, May 4, 2021.

The meeting is hosted by DataStax Developers as part of KubeCon Europe.

About DataStax & K8ssandra: On November 18, 2020, DataStax released K8ssandra at the KubeCon + CloudNativeCon North America Conference, an open source, new release version of Apache Cassandra deployed on Kubernetes.K8ssandra allows developers to complete the deployment of Cassandra with only a Kubernetes Helm chart. K8ssandra is now available for download and use. It allows database administrators (DBA) and website reliability engineers (SRE) to flexibly increase or decrease data on Kubernetes.


Relate to Course


We mentioned K8S and Docker in Config management course slide, I would like to know more about that. I think this meeting will help to learn how these kinds cloud-native system handle database and applications.


Details

In many distributed systems(including my research project) building highly scalable, cloud-native applications, have struggled with finding a data solution that can keep pace with the needs of their application. Often this has meant a choice between using a managed database, or taking on the task of deploying and managing a distributed database on Kubernetes. Apache Cassandra is a powerful database used in many applications that require resilience and scalability, but has traditionally been challenging to deploy and manage effectively on Kubernetes. Now the Kubernetes and Cassandra communities have come together in a great collaboration called K8ssandra, a production-ready platform that makes it easy to run Cassandra on Kubernetes.

in the meeting, they included below items:

  *   Basic idea about K8ssandra
  *   Creating a K8ssandra release
  *   Monitoring database performance through metrics and alerts
  *   Breaking nodes in the database cluster on purpose to observe how Cassandra self-heals
  *   ...

why I choose it

I'm working on building a distributed GPU ML platform, k8s and Docker talked in class gives me inspire that we may use these things to help us manage data(basically sparse matrix and graph store in GPU global memory )


what I learned

Frankly, I didn't get all of the knowledge that I expect. Because I'm more focused on understanding basic information and knowledge about K8ssandra but less concerned about how to use it. They only mentioned the basic idea about it and use the rest of the time to teach me how to install and use their product. Fortunately, they also provide another video((6) DoKC Day: Eric Zietlow & Aleks Volochnev "Deploying open cloud-native data using K8ssandra" - YouTube<https://www.youtube.com/watch?v=7WfN9P4mYTQ>) that talks more details about database structure, how data is distributed, and how data is replicated between nodes.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.stevens.edu/pipermail/cs615asa/attachments/20210506/34936adf/attachment.html>