An Introduction To Cassandra

Introduction to Apache Cassandra (September 2014). Design principles, replication, consistency, clusters, CQL.

A short introduction to replication and consistency in the Cassandra distributed database. Delivered April 28th, 2010 at the Seattle Scalability Meetup.

Cassandra is a distributed database management system designed to handle large amounts of data across many commodity servers. It provides high availability with no single points of failure and linear scalability as nodes are added. Cassandra uses a peer-to-peer distributed architecture and tunable consistency levels to achieve high performance and availability without requiring strong consistency. It is based on Amazon's Dynamo and Google's Bigtable papers and provides a combination of their features.

Apache Cassandra is a free, distributed, open source, and highly scalable NoSQL database that is designed to handle large amounts of data across many commodity servers. It provides high availability with no single point of failure, linear scalability, and tunable consistency. Cassandra's architecture allows it to spread data across a cluster of servers and replicate across multiple data centers for fault tolerance. It is used by many large companies for applications that require high performance, scalability, and availability.

I don't think it's hyperbole when I say that Facebook, Instagram, Twitter & Netflix now define the dimensions of our social & entertainment universe. But what kind of technology engines purr under the hoods of these social media machines? Here is a tech student's perspective on making the paradigm shift to "Big Data" using innovative models: alphabet blocks, nesting dolls, & LEGOs! Get info on: - What is Cassandra (C*)? - Installing C* Community Version on Amazon Web Services EC2 - Data Modelling & Database Design in C* using CQL3 - Industry Use Cases

The Wikimedia Foundation is a non-profit and charitable organization driven by a vision of a world where every human can freely share in the sum of all knowledge. Each month Wikimedia sites serve over 18 billion page views to 500 million unique visitors around the world. Among the many resources offered by Wikimedia is a public-facing API that provides low-latency, programmatic access to full-history content and meta-data, in a variety of formats. Commonly, results from this system are the product of computationally intensive transformations, and must be pre-generated and persisted to meet latency expectations. Unsurprisingly, there are numerous challenges to providing low-latency storage of such a massive data-set, in a demanding, globally distributed environment. This talk covers Wikimedia Content API, and it's use of Apache Cassandra, a massively-scalable distributed database, as storage for a diverse and growing set of use-cases. Trials, tribulations, and triumphs, of both a development and operational nature are discussed.

This document summarizes Eric Evans' presentation on using Cassandra as the backend for Wikimedia's content API. It discusses Wikimedia's goals of providing free knowledge, key metrics about Wikipedia and its architecture. It then focuses on how Wikimedia uses Cassandra, including their data model, compression techniques, and ongoing work to optimize compaction strategies and reduce node density to improve performance.

Among the resources offered by Wikimedia is an API providing low-latency access to full-history content, in many formats. Its results are often the product of computationally intensive transforms, and must be pre-generated and stored to meet latency expectations. Unsurprisingly, there are many challenges to providing low-latency access to such a large data-set, in a demanding, globally distributed environment. This presentation covers the Wikimedia content API and its use of Apache Cassandra as storage for a diverse and growing set of use-cases. Trials, tribulations, and triumphs, of both a development and operational nature will be discussed.

This document discusses using Apache Cassandra to store and retrieve time series data more efficiently than the traditional RRDTool approach. It describes how Cassandra is well-suited for time series data due to its high write throughput, ability to store data sorted on disk, and partitioning and replication. The document also outlines a data model for storing time series metrics in Cassandra and discusses Newts, an open source time series data store built on Cassandra.

This document discusses using Apache Cassandra to store and manage time series data in OpenNMS. It describes some limitations of the existing RRDTool-based data storage, such as high I/O requirements for updating and aggregating data. Cassandra is presented as an alternative that is optimized for write throughput, flexible data modeling, high availability, and ability to perform aggregations at read time rather than write time. The Newts project is introduced as a standalone time series data store built on Cassandra that aims to provide fast storage and retrieval of raw samples along with flexible aggregation capabilities.

Whether it's statistics, weather forecasting, astronomy, finance, or network management, time series data plays a critical role in analytics and forecasting. Unfortunately, while many tools exist for time series storage and analysis, few are able to scale past memory limits, or provide rich query and analytics capabilities outside what is necessary to produce simple plots; For those challenged by large volumes of data, there is much room for improvement. Apache Cassandra is a fully distributed second-generation database. Cassandra stores data in key-sorted order making it ideal for time series, and its high throughput and linear scalability make it well suited to very large data sets. This talk will cover some of the requirements and challenges of large scale time series storage and analysis. Cassandra data and query modeling for this use-case will be discussed, and Newts, an open source Cassandra-based time series store under development at The OpenNMS Group will be introduced.

OpenNMS User Conference Europe presentation on using Apache Cassandra and Newts for time-series data storage.

Presented at Cassandra London (April 7, 2014); The challenges of time-series storage and analytics in OpenNMS, with an introduction to Newts, a new Cassandra-based time-series data store.

A presentation on the recent work to transition Cassandra from its naive 1-partition-per-node distribution, to a proper virtual nodes implementation.

This document summarizes a presentation about modeling data with Cassandra Query Language (CQL) using examples from a Twitter-like application called Twissandra. It introduces CQL as an alternative to Thrift for querying Cassandra and describes how to model users, followers, tweets, timelines and other social media data structures in Cassandra tables. The presentation emphasizes denormalizing data and using materialized views to optimize queries, and concludes by noting that applications can be built in various languages thanks to Cassandra drivers.

CQL is the query language for Apache Cassandra that provides an SQL-like interface. The document discusses the evolution from the older Thrift RPC interface to CQL and provides examples of modeling tweet data in Cassandra using tables like users, tweets, following, followers, userline, and timeline. It also covers techniques like denormalization, materialized views, and batch loading of related data to optimize for common queries.

The document discusses topology and partitioning in Cassandra distributed hash tables (DHTs). It describes issues with poor load distribution and data distribution in traditional DHT designs. It proposes using virtual nodes, where each physical node is assigned multiple tokens, to better distribute partitions and improve performance. Configuration options for Cassandra are presented that implement virtual nodes using a random token assignment strategy.

The document discusses Cassandra's topology and how it is moving from a single token per node model to a virtual node model where each node is assigned multiple tokens. This improves load balancing and data distribution in the cluster. Specifically, it addresses problems with the single token approach like poor load distribution when nodes fail and inefficient data movement when adding or replacing nodes. The virtual node model with random token assignment provides better scaling properties as the number of nodes and data size increases.