Comparing Graph Databases
source link: https://www.tuicool.com/articles/6ZJR7f6
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
Part 2: ArangoDB, OrientDB, and AnzoGraph DB
It’s great to see how passionate people are about their favorite graphing database provider or, at least, the graphing database company they work for. Since there are so many available options, please check out db-engine’s list of 33 different options for your graphing database needs. In this article, I will briefly highlight:
- OrientDB
- ArangoDB
- AnzoGraph DB
These three rank 3rd, 4th, and 26th in popularity, according to the list provided by db-engines.com . In the interest of credible journalism, I will humbly attempt to be as unbiased as possible, and report on the information provided on the company’s website as well as recent articles.
Please see my previous article (Part 1) for a quick comparison of relational database management systems and graph databases.
OrientDB — “The database designed for the modern world”
Website: https://orientdb.com/
Documentation: https://orientdb.org/docs/3.0.x/
Initially released in 2010 , OrientDB supports many programming languages including: .Net, C, C#, C++, Clojure, Java, JavaScript, JavaScript(Node.js), PHP, Python, Ruby, and Scala. OrientDB is a schema-free multi-model database system supporting graph , document, key/value, and object models. It supports schema-less, schema-full, and schema-mixed modes. Gremlin and SQL queries are both supported for graph traversal. OrientDB is implemented with Java, so it can be run on all operating systems with a Java JDK ≥ JDK6. It is open-sourced, but commercial support is available from OrientDB as well.
According to their website,OrientDB provides the service of a graph database system without the need to “deploy multiple systems to handle other data types.” This method serves to increase “performance and security while supporting scalability.” OrientDB differentiates itself from the many graph database systems by managing a multi-model system by design. It does not simply “add layers for additional models, resulting in decreased performance.”
There are 33 user reviews on G2 with an average rating of 4/5 stars. The majority of the reviews are very positive, and it should be noted that the last average review was from June 2016, so it seems like OrientDB is doing a stand-up job of fixing bugs and deploying fully-developed features. The main criticism of reviews seems to be the desire for more robust documentation. Main praises include the reasonable price, quick-installation, and that it is user-friendly.
ArangoDB — “One engine. One query language. Multiple models.”
Website: https://www.arangodb.com/
Documentation: https://www.arangodb.com/documentation/
Originally called AvocadoDB in 2011, as evident in its logo, ArangoDB came into being in 2012 . ArangoDB is open-sourced, multi-model (key/value, documents, and graphs), and implemented with C, C++, and JavaScript. Server operating systems include: Linux, OS X, Raspbian, Solaris, and Windows. It is schema-free and supports the following languages: C#, Clojure, Java, JavaScript (Node.js), PHP, Python, and Ruby. ArangoDB operates with one database core and its own unified query language AQL (ArangoDB Query Language), which is similar to SQL in many ways. AQL is declarative and allows the combination of different data access patterns in a single query. ArangoDB was designed specifically to allow key/value, document, and graph data to be stored together and queried with a common language.
According to their website , ArangoDB can operate as a distributed & highly scalable database cluster. It runs on Kubernetes, including persistent primitives & easy cluster setup. ArangoDB has natively integrated cross-platform indexing, text-search and ranking engine for information retrieval and it is optimized for speed and memory. Full GeoJSON support is also provided.
There are 41 user reviews on G2 with an average of 5/5 stars. The single average rating from 2017 indicated a lack of SQL support and dissatisfaction of having to adapt to AQL. Others refer to AQL as “intuitive” and describe ArangoDB as “feature-rich.”
For a sneak peek at the new features in the upcoming 3.6 release, check out ArangoDB’s Webinar on Oct 10th, 2019 at 1PM EST.
AnzoGraph — “Build Your Solutions on a Fast, Scalable Database”
Website: www.anzograph.com
Documentation: https://docs.cambridgesemantics.com/anzograph/userdoc/home.htm
Initially released in 2018, this commercial graph database operates with the RDF (Resource Description Framework). The RDF model represents information as triples in the form of subject-predicate-object. RDF stores can be considered as a subclass of graph DBMs, but RDF stores are distinguished as they offer specific methods going beyond the general graph DBMS. Most RDF stores, AnzoGraph included, support SPARQL, which is a SQL-like query language used for OLAP(Online analytical processing) -style analytics. AnzoGraph DB’s operating server is Linux and it supports C++ and Java.
According to their website,AnzoGraph DB is built for online data analytics with performance that linearly scales. It is a Massively Parallel Processing (MPP) native graph database built for analytics at scale (trillions of triples and more), speed and deep link insights. It is intended for embedded analytics that require graph algorithms, graph views, named queries, aggregates, built-in data science functions, data warehouse-style BI, and reporting functions. You can try out their real-world test, check out their benchmark study , and download a 60-day free trial .
Some reviews of AnzoGraph DB exist right here on Medium. Check out this article: Graph Databases. What’s the Big Deal? byFavio Vázquez. He points out that “graph OLAP databases are becoming very important as Machine Learning and AI grows since a number of Machine Learning algorithms are inherently graph algorithms and are more efficient to run on a graph OLAP database vs. running them on a RDBMS.”
Another examination of AnzoGraph exists in the article written by George Anadiotis , which compares AnzoGraph to TigerGraph.
Conclusion
There are many options for graphing databases, and it seems as though each is trying to find their own personal corner of the market. Which one is the best? It really depends on your needs. Every Graph DB will have its unique strengths, weaknesses, and benchmarks. And, as these systems develop and grow their weaknesses will change, and they will likely become more comprehensive and capable. Take the time to shop around and educate yourself on all the available options because there are a lot, and the numbers keep growing.
Recommend
About Joyk
Aggregate valuable and interesting links.
Joyk means Joy of geeK