December 2016 Summaries

The Neo4j blog features a collection of articles, podcasts, and code repositories related to graph databases. Articles cover various topics such as data migration, integration with other tools like Apache Zeppelin, and using Neo4j for different applications like fraud analysis and sports data visualization. The blog also showcases user-generated content, including a "From the Community" section where featured posts are highlighted, and a sign-up form to encourage users to share their own contributions. Additionally, podcasts feature interviews with experts in the field, such as Evelina Gabasova from the University of Cambridge, while code repositories demonstrate various Neo4j-related projects and libraries, including examples of using the JDBC driver and Docker.

The Neo4j team has started the release cycle for version 3.2, with a planned final version expected to be released in the first half of 2017. The new release cycle will follow a more iterative approach, with alpha releases every two weeks, followed by beta releases after feature freeze, and then release candidate versions before the general availability (GA) release. The team is also working on improving the "compiled" runtime for Cypher, which will benefit execution performance and memory efficiency, especially on large queries. Notable changes in this release include the removal of the RULE planner in Cypher, introduction of a lock acquisition timeout, and use label scan store by default for schema index population. The team invites users to provide feedback through various channels, including email, GitHub, Discord, and Google Groups.

Doug Balog, a business leader at IBM Power Systems, discusses the changing landscape of computing hardware and its impact on the software industry. He explains how IBM's approach to hardware innovation is shifting towards a new open-source framework that benefits graph technology ecosystems like Neo4j. Balog highlights the marriage between Neo4j and IBM Power Systems as a perfect intersection, enabling scalable platforms for graph databases. He also discusses the demise of Moore's Law and its implications on processor technology, predicting a future where innovation occurs through acceleration technologies such as flash memory, I/O, FPGAs, and GPUs. The OpenPOWER Foundation, an open-source initiative co-founded by IBM, is another key aspect of this shift, bringing together industry leaders to develop new hardware innovations. Balog emphasizes the need for companies to disrupt themselves in order to stay relevant, highlighting IBM's own transformation as a prime example.

Ivan Portilla, an IT architect at IBM, discusses cognitive computing and its applications in various industries. He explains that cognitive computing is a field of research that focuses on creating machines that can understand human language, reason, and learn from experience. Ivan shares examples of how IBM's Watson system has been used to detect cancer, provide healthy food recommendations for hotel guests, and streamline the travel industry. He also highlights the importance of POWER8 hardware architecture in supporting cognitive computing applications. Additionally, Ivan emphasizes the need for businesses to understand cognitive computing and use the right tools and resources to implement it effectively.

Neo4j 3.1 introduces a new state-of-the-art clustering architecture called Causal Clustering, which enables support for ultra-large clusters and a wider range of cluster topologies. This new release also includes built-in load balancing, causal consistency, multiple users, role-based access control, query and security event logging, list and terminate running queries, long-requested improvements in the database kernel, and new command-line interfaces. Additionally, Neo4j 3.1 supports IBM POWER8 with CAPI Flash, which enables ultra-high scalability and in-memory graph processing on an ultra-large scale. The release is designed to meet enterprise requirements for scale and security, providing a comprehensive set of graph functionality and building upon the advanced client drivers introduced in Neo4j 3.0.

Neo4j 3.1 introduces a new state-of-the-art clustering architecture using the Raft protocol, enabling support for ultra-large clusters and wider cluster topologies, including built-in load balancing and causal consistency with tunable clustering consistency models. The release also includes enterprise-class security features such as multiple users, role-based access control, query and security event logging, list and terminate running queries, fine-grained access control, and a new schema viewer for graph model visualization. Additionally, Neo4j 3.1 introduces long-requested improvements in the database kernel, including more efficient space management and execution guard for max query duration per instance. The release also supports IBM POWER8 with CAPI Flash, enabling ultra-high scalability for graph processing on an ultra-large scale.

FactGem, a company that builds data visualization tools, uses Neo4j as its underlying graph database to create a "data fabric" for companies to maintain and visualize their existing investments. The company's CEO, Megan Kvamme, notes that Neo4j's ability to expose data connections quickly has been crucial in allowing FactGem to deliver results to customers rapidly. Kvamme attributes the success of their partnership to Neo4j's willingness to work closely with its partners to achieve mutual success. The partnership has enabled FactGem to move at fast speeds and deliver results to enterprise customers, which is particularly important for a company that operates in real-time. Overall, Kvamme expresses her satisfaction with the relationship and praises Neo4j for its collaborative approach and commitment to helping its partners succeed.

The Neo4j Movie App Template provides a foundation for building a Neo4j project or experiment using Node.js or React.js. It allows users to log in, rate movies, and receive movie recommendations. The template includes an authentication system that uses passwords and API keys. Users can create accounts, log in, and rate movies. The app also recommends movies based on user ratings and keywords. The recommendation engine uses a combination of methods, including user-centric, user-based, movie-centric, and keyword-based approaches.

The Neo4j-Drupal module is a powerful integration that enables non-developers to work with graph databases using a graphical user interface. It combines the content modeling capabilities of Drupal with the data modeling features of Neo4j, allowing users to create complex systems without extensive coding knowledge. The module has been re-implemented under a separate namespace and uses the Bolt PHP driver for improved performance. It supports both Drupal 7 and 8 versions, but the Drupal 8 version is currently more feature-rich due to its richer APIs. The module provides a simple way to integrate Neo4j with Drupal's Rules module, enabling site builders to define actions that can be triggered by events without writing code. With this integration, users can explore graph databases using a GUI and take advantage of the capabilities offered by Neo4j in the Drupal ecosystem.

Neo4j 3.1 is becoming production ready on December 13, 2016, and a live launch event has been scheduled for that date to showcase the new features of Neo4j 3.1. The event will feature CEO Emil Eifrem and other members of the Neo4j Product team, providing an overview of the graph space and a comprehensive understanding of Neo4j 3.x series. Highlights of Neo4j 3.1 include a new Causal Clustering architecture for large clusters, improved security features, user-defined functions, a new Cypher command-line interface, and enhancements to the Neo4j Browser. The event will provide an in-depth look at these features, as well as conclude with a review of the product roadmap.