February 2019 Summaries

Apache Cassandra, an open source distributed NoSQL database, has gained popularity for its high availability, performance, and scalability in enterprise applications. However, the demand for skilled engineers to manage this technology is growing faster than the supply of qualified experts. To ensure successful deployment of Cassandra, organizations should invest in training their teams on the technology, provide additional resources for self-paced learning, carefully choose data models, and consider using production-ready implementations like DataStax Distribution of Apache Cassandra to manage costs and complexity as they scale.

In this post, the author discusses the development of a new implementation of microservices for KillrVideo in Python. They start by defining service interfaces using Google Remote Procedure Call (GRPC) and generating stubs for Python services. The author explains why GRPC was chosen over other options like RESTful APIs or GraphQL, highlighting its advantages such as consistent semantics across multiple language implementations. They also demonstrate how to create bindings in Python using grpcio-tools package and discuss the process of implementing each service by extending server stubs generated by GRPC. Finally, they showcase an example of starting a GRPC server and registering services with it.

Consistent hashing is a type of hashing algorithm that maps data to variable length and provides an effective pattern for mapping keys to particular nodes in distributed database systems like Apache Cassandra or DataStax Enterprise. This method minimizes reorganization when nodes are added or removed, with partitions based on partition keys. The partition key should not be confused with a primary key; it is more like a unique identifier controlled by the system that makes up part of a composite key. In a four-node cluster, data distribution occurs according to the value of the partition key and the range each node is responsible for.

The transition to 5G networks is expected to significantly increase mobile use and wireless device connectivity, including robots, security cameras, and smart cars. This will require new hybrid cloud data management strategies that can handle larger volumes of data and meet higher standards of reliability and performance. Industries such as healthcare, manufacturing, automotive, energy production, and media and entertainment are expected to undergo significant transformations due to 5G technology. Organizations must prepare for the "5G data crush" by implementing and maintaining a data management strategy that can accommodate larger volumes of data while maintaining high speeds and eliminating the risk of downtime. This will involve reviewing current database capabilities, using distributed cloud-based data layers, and adopting efficient, scalable, and secure solutions like DataStax Enterprise to manage customer demands effectively.

The author took a break from travel to work on a project they had been contemplating for some time - creating an implementation of the KillrVideo reference application in Python, specifically focusing on the microservices tier. They discuss their reasons for choosing this project and share their experiences so far. They also mention that Amanda Moran has agreed to collaborate with them on this project and welcome contributions from anyone interested in helping out.

The DataStax Developer Relations team has observed a growing trend in the use of Apache Cassandra among developers for its ability to provide scale, performance, uptime, and distribution. Many industries are utilizing Cassandra for various applications such as financial data, customer data, IoT systems, logging, and metadata storage. However, challenges arise with hybrid-cloud and multi-cloud deployments, including latency, security, and administration issues. Database security remains a significant concern among developers due to the increasing number of high-profile data breaches. DataStax Enterprise offers advanced security features to address these concerns. The team plans to focus on application development topics in their upcoming series and blog posts.

Sling TV, an over-the-top live streaming content platform, aims to provide a highly resilient service that is personalized for each user and can scale on demand to accommodate its growing customer base. The company has chosen DataStax Enterprise (DSE) as part of their hybrid cloud strategy due to its ability to support their next-gen architecture goals and enable data replication across the country in less than two seconds. Sling TV is currently the number one live TV streaming service with approximately 2.3 million customers, and they plan to leverage DSE for future growth.

DataStax emphasizes hiring innovative, collaborative, and hard-working individuals from a global pool of candidates due to its highly distributed workforce. With over 65% of employees being fully remote, this approach allows for a diverse team that is featured in the "Humans of DataStax" video series. The first two videos highlight HR Specialist Lynsey Reys-Nickel and Developer Advocate Amanda Moran, who exemplify the company's core behaviors of accountability, visibility, respect, and growth mindset. Both employees contribute to creating a positive work environment and improving DataStax products for users and developers alike.

Valentine's Day brings significant retail sales, with an estimated $19.6 billion spent in 2018 alone. Retailers and database administrators face challenges managing the surge in online traffic during this period. To ensure smooth operations and a positive customer experience, companies need to focus on scalability, hybrid-cloud applications, and high availability through masterless architecture databases. NoSQL databases can help scale applications by adding nodes without downtime or maintenance issues, while hybrid cloud strategies streamline data management and improve accessibility during peak traffic times.