December 2016 Summaries

InfluxData experienced a productive and hectic year in 2016, marked by significant team additions, business model establishment, and major software releases. The company released several versions of its core products, including InfluxDB, Telegraf, Chronograf, Kapacitor, and Cloud offerings, as well as announcing a Series B funding round led by Battery Ventures. Key milestones included the release of the 1.0 version of the TICK stack and InfluxDB Enterprise, an all-new version of Chronograf, and major updates to query engines and storage systems.

In the spirit of new year's resolutions, InfluxData has put together some advice for working with InfluxDB, maintaining a healthy database, and discovering new tools like Chronograf. Downsampling high precision data can improve query performance, eliminate storage concerns, and enhance graphing tool performance by removing unnecessary data from the database. Focusing on series cardinality is crucial to maintain database health, as extremely high cardinality can kill the InfluxDB process, and strategies such as monitoring, managing, and setting limits can help mitigate this issue. Chronograf offers an application-based view of infrastructure, data visualizations, and a user interface for Kapacitor, simplifying the monitoring solution and allowing users to get started with a free trial. Additionally, InfluxData is offering downloads for the TICK-stack, deploying on servers or cloud platforms, and sharing customer testimonials.

The InfluxDB team recently announced several new features, including multi-tenant support for Grafana on InfluxDB Cloud, which allows customers to create dedicated instances of Grafana with configurable access to different groups and users. InfluxDB has been successfully used by BBOXX, a company that sells solar home systems in rural Africa, to store data from hundreds of thousands of units while maintaining low latency for querying individual units. The InfluxDB community is growing, and the team is seeking featured applications or products using InfluxDB in production, with a "thank you" gift of an InfluxDB hoodie and stickers for those who participate.

In this week's recap, the InfluxDB community discussed various issues with field keys, queries, and Continuous Queries (CQs). Some common explanations for missing data include retention policies, query time ranges, and schema inconsistencies. The community also explored advanced CQ syntax to configure the time range of a CQ, ensuring it captures all relevant data points. Additionally, InfluxDB has released new downloads for the TICK-stack, cloud deployments, server-side trials, and is encouraging users to share their success stories.

** The technology industry is expected to experience significant growth in various areas, including server-side programming languages, cloud computing, and artificial intelligence. Go is predicted to continue its rise as a dominant server-side language, with improvements in performance and tooling expected with the release of Go 1.8. Kubernetes is also expected to maintain its popularity, with major industry players and new startups lining up behind it. Serverless computing is anticipated to gain traction, but will face challenges related to developer productivity and usability. AI and deep learning hype are predicted to continue, but may not deliver on their promises due to the lack of fundamental understanding of intelligence and learning. Mixed reality (MR) and virtual reality (VR) hype is expected to reach its peak before heading into a trough of disillusionment, while IoT will continue to be fragmented but see an exponential expansion of sensor data. Finally, 2017 is predicted to be an exciting year for the tech industry, with many opportunities for software developers.

The Chronograf beta has been updated with bug fixes and new features, including the ability to create custom queries outside of the query builder, improved UI enhancements such as distinguishing queries in Data Explorer and color-coding Kapacitor alert levels. Additionally, a couple of bug fixes were included, addressing issues like text selection and version information display. The release also includes other improvements and updates, with more details available in the change log and readme.

InfluxDB has made significant announcements, including adding multi-tenant support for Grafana on InfluxDB Cloud, and introducing Chronograf, a complete open-source monitoring solution running on the TICK stack. The company also highlighted customer success stories, such as Coupa's implementation of dynamic monitoring using InfluxDB, Kafka, and Grafana. Additionally, InfluxDB has been featured in various benchmarks, including a comparison with OpenTSDB, which showed InfluxDB outperforming its competitor. Furthermore, the company is hiring and is offering a "thank you" gift to customers who feature their use of InfluxDB on the official website.

InfluxDB Tech Tips - Understanding Dependent Tags In Series Cardinality InfluxDB supports writing historical data using the line protocol, with optional timestamp specification in Unix time format, allowing for flexible and efficient data storage. Timestamps have a maximum allowable value of 9223372036854775806 and minimum value of -9223372036854775806, and can be specified with precision using the HTTP API's `precision` query string parameter. Series cardinality is determined by the number of unique tag sets for a single measurement, not just the product of tag values per tag. When querying data with disparate time ranges in the WHERE clause, InfluxQL does not currently support it due to time gaps, and users must split their queries into separate queries for each range.

InfluxDB Week in Review Dec 12, 2016 The InfluxDB community has seen several recent announcements, including new DC/OS packages, a blog post on agile testing with Kubernetes and Heapster, and the release of Chronograf, a complete open source monitoring solution running on the TICK stack. Additionally, InfluxDB has been compared favorably to OpenTSDB in a benchmark review, and customers such as Wrecking Solutions have reported positive experiences with the product for diagnostic and error logging in their SaaS products. The company is also offering free virtual training sessions, including an "Intro to Kapacitor for Alerting and Anomaly Detection" session, and has announced the addition of multi-tenant support for Grafana on InfluxDB Cloud, allowing customers to configure access to different groups and users. Furthermore, InfluxData is seeking talented developers to join their team, with a call to claim an InfluxDB hoodie as a "thank you" gift for featuring a company's use of InfluxDB in production.

In this week's recap of InfluxDB timestamps and TICK-stack related issues, functions with an INTO Clause were discussed. InfluxDB supports negative timestamps, but when using the DISTINCT() function with an INTO clause, it may overwrite duplicate points in the destination measurement instead of returning them. Additionally, the COUNT() function with fill() can be used to replace null values with a given fill_option, which is useful for handling time intervals with no data. The TICK-stack's downloads are now live, and InfluxDB Cloud and Enterprise trials are available for free.

InfluxDB Enterprise's write performance is evaluated in this benchmark test, which assesses the impact of various parameters such as number of data nodes, replication factor, and batch size on throughput. The results show that InfluxDB Enterprise scales horizontally for writes, achieving higher write throughput with more data nodes, lower write throughput with a higher replication factor, and higher write throughput with larger batch sizes. A 32-node cluster can handle over 11 million singly-replicated writes per second and 9.3 million doubly-replicated writes per second, outperforming single nodes of InfluxDB on the most performant hardware. The benchmark test aims to provide a consistent comparison of InfluxDB and InfluxDB Enterprise, with plans for periodic updates and further analysis.

Chronograf, the user interface of the TICK stack, has been released in beta status after a month of rapid feature iteration based on user feedback. The new version includes OAuth authentication via GitHub, application templates for various platforms, and a responsive design for the host view page. A video walkthrough is also available to introduce users to the key capabilities of Chronograf. The project aims to drive towards general availability, with plans to release nightly builds every two weeks and address community feedback on features such as an application-based view of infrastructure, export of graphs, and user-configurable dashboards. InfluxData's vision for Chronograf is to enable developers to own their monitoring using pure open-source software, optimizing for developer happiness and delivering tools that are a joy to use.

InfluxDB has recently announced new features, including multi-tenant support for Grafana in InfluxDB Cloud, allowing customers to configure access to different groups and users. The company is also hiring and offering free hands-on workshops and training sessions on various topics related to InfluxDB. Several customer testimonials highlight the ease of use and effectiveness of InfluxDB in creating detailed application performance monitors and other monitoring solutions. Additionally, InfluxDB has been featured in a benchmark comparison with OpenTSDB, where it outperformed its competitor. The company is also giving away free hoodies to customers who feature their applications or products on influxdata.com.

InfluxDB has introduced regular expressions in the SELECT clause, allowing users to specify multiple field keys without listing them all out. Additionally, InfluxQL does not currently support dropping specific field keys from a measurement, but users can work around this limitation by writing the desired field keys to a different measurement, dropping the original measurement, and then writing the saved field keys back to the original measurement. The selector function in InfluxQL returns a single point that includes the last value of one or more specified fields, with the timestamp being InfluxDB's null timestamp equivalent when multiple timestamps are involved.