January 2018 Summaries

Data Day Texas premiered a new Apache Cassandra™ track, featuring presentations on data modeling, use cases, and operations. The event attracted 1000 attendees, making it the largest Data Day yet. In addition to Cassandra talks, there were tracks on Apache Spark™, Graph, streaming technology, and AI. Highlights included Jeff Carpenter's deep dive into Cassandra architecture, Ben Bromhead's discussion of deploying Cassandra with Kubernetes, Jon Haddad's presentation on optimizing performance in a Cassandra cluster, Dikang Gu and Pengchao Wang's update on experimental work at Facebook, Aaron Ploetz's insights on performance data modeling, and Jonathan Ellis's talk on Apache Cassandra and the cloud. The community is looking forward to an even bigger Data Day event in San Francisco later this year.

The cloud has revolutionized customer engagement by enabling enterprises to use data from various touchpoints to create personalized experiences. However, customers are not concerned about the underlying processes and security measures. DataStax Enterprise, built on Apache Cassandra NoSQL database, provides a seamless experience for global enterprises in different industries. McDonald's, eBay, Sony, UBS, Walmart, and others rely on DataStax to deliver impactful digital experiences while protecting their data. The partnership with Oracle Cloud Infrastructure aims to enhance the ability to power real-time customer experiences for businesses seeking success in today's fast-paced economy.

CES 2018 showcased the shift from gadgets being the stars to the technology behind them, such as AI and IoT. Real-time data and insights are making today's devices smarter and more adaptable to users' needs. The focus is now on software over hardware, with real-time, AI, and IoT playing crucial roles in shaping the future of technology.

Recent security vulnerabilities Meltdown and Spectre pose threats to sensitive information exposure and have led to fixes in operating systems and CPUs that may affect the performance of DataStax Enterprise. The overall performance impact on Linux distributions patched for Meltdown and Spectre was usually below 5% and did not exceed 10%. To determine the impact for your production workload, you will need to run tests in your performance test environment with a workload that matches your production workload.

This article discusses Gremlin query optimization for DSE Graph, focusing on identifying bottlenecks and measuring improvements to achieve performant real-time transactional Gremlin queries. It highlights common mistakes in writing Gremlin queries and provides techniques to avoid them. The article also covers the use of as/select steps, where step, match step, .or condition, paging with range step, and using search index for paging. Additionally, it mentions the timeLimit step for speeding up query iterations during troubleshooting. As DSE Graph's optimizer matures, some manual optimizations may become unnecessary in the future.

In this three-part series, we will discuss upgrading the KillrVideo Java reference application from Cassandra based tabular searches to using DSE Search. Part 1 focuses on the "before" picture and how searches were previously implemented with Cassandra tables. We explore various types of searches in KillrVideo, including typeahead search, tag-based search, and more videos like this search. The text also delves into the underlying CQL schemas for tags_by_letter and videos_by_tag tables, as well as the VideoAddedHandlers class responsible for performing actions every time a video is added to KillrVideo. In part 2, we will explain the transition to DSE Search and considerations taken into account, along with before and after code comparisons. Finally, in part 3, we will examine our results and explore more advanced types of searches that can now be performed.