January 2019 Summaries

InfluxData experienced accelerated growth in 2018, securing $35 million in Series C financing, maintaining its leadership position among time series databases, and doubling sales to 450 customers across DevOps and IoT functions. The company's open source community grew significantly, with over 192,000 active instances and 15k stars on Github. InfluxData also made significant strides in product development, including performance, data compaction, machine learning systems, LDAP integration, and advanced analytics. With a strong team and momentum, the company is poised for another transformative year in 2019, with plans to introduce significant product updates and expand its reach to more people and organizations.

InfluxData's Gianluca Arbezzano highlights the importance of observability in understanding complex systems, particularly distributed systems. Monitoring is just a starting point, and true observability requires tools, methodologies, and a mindset to learn about a system. Distributed tracing is an essential tool for observability, allowing teams to understand how requests flow through multiple services and applications. By using distributed tracing, Arbezzano was able to identify the root cause of a failure in his company's cluster creation process, which involved over 40 interactions with AWS. The ability to visualize and query traces provides valuable insights into system behavior, making it easier to troubleshoot and improve the development cycle. Observability is not just important for production but also for development, where it can help improve feature development speed and accuracy.

InfluxData has released InfluxDB 2.0 alpha, aiming to unify the TICK Stack into a single cohesive API, incorporating time series database, UI and dashboarding tool, and background processing and monitoring agent. The development process will progress through alpha, beta, and general availability phases. Paul Dix, founder and CTO of InfluxData, discusses the planned release in detail, highlighting Flux as a key feature, a new data scripting and query language. The company invites community feedback on this early release, seeking input to refine the product before its official launch.

The Manufacturing Tomorrow Q&A with Navdeep Sidhu, Head of Product Marketing at InfluxData, discusses how manufacturing companies are leveraging data to create real business value. Sidhu provides an overview of InfluxData and highlights the pressing data-related issues faced by the industry, including measurement, monitoring, and access to real-time data. He also addresses barriers to adopting data analysis in the manufacturing sector, which includes challenges such as limited technical expertise and high costs associated with implementing new technologies. Additionally, Sidhu shares InfluxData's expansion plans in the global manufacturing arena, indicating a growing focus on supporting manufacturers in harnessing the power of data-driven decision making.

InfluxDB Enterprise 1.7.3 has been released with several new features and improvements, including a technical preview of Flux, a new LDAP configuration setting for secure connections, enhanced TLS configuration options, and additional defect fixes from previous versions. The Anti-Entropy service is now off by default to ensure better performance in production environments. Customers can download the binaries directly from the InfluxDB Enterprise portal.

I created the http_listener_v2 Telegraf input plugin to monitor solar panels because I wanted more access to how much power and energy my solar panels were generating, as well as correlate this data with temperature, cloudiness, and sun intensity. I was using Fronius solar inverter which sends HTTP POST requests with JSON encoded data to an endpoint, but the InfluxDB metrics collection agent Telegraf didn't support parsing JSON data at that time. Therefore, I created a plugin to extend its capabilities and persist the inverter-produced data in InfluxDB.

The new version of InfluxDB, version 2.0, is now available in alpha access to the public, offering a simplified setup experience and all the necessary components in a single binary. It includes features such as native Prometheus metric scraping, improved Telegraf configuration options, and a new Data Explorer for quick data visualization with Flux. The platform also introduces tasks for scheduling queries on a schedule and storing results, a unified API for interacting with the system, and a command-line interface for scripting queries. Users can authenticate as users, set up organizations and buckets, and upload raw line protocol files to store their time series data. The software is designed to be community-driven, with plans for regular updates and feedback mechanisms to ensure it meets the standards of developers and users alike.

InfluxDB 2.0 is a significant evolution of the InfluxData platform, aiming to collapse the TICK Stack into one cohesive whole by combining time series database, UI and dashboarding tool, and background processing and monitoring agent behind a single API. The new language, Flux, is designed to support graphical user experiences, integrate across disparate data sources, and be Turing complete for querying and working with not only time series data but also general data. Flux has been developed as a separate repository from InfluxDB, using the liberal MIT license, and will have a life of its own as an open source project useful whether used with or without InfluxDB. The TICK stack is being unified into one cohesive whole, with the goal of making InfluxDB a multi-tenanted, time series service. InfluxDB 2.0 includes first-class support for the Prometheus exposition format and both push and pull models out of the box. Future work includes adding compatibility with older APIs, features for backup and restore, bulk data import and export, and data deletes. The alpha release is available now, with weekly versioned builds starting in February, followed by a beta phase with operational readiness activities.

Factry, a Belgian software company, brings open software to the manufacturing industry through its services and products, including Factry Historian, which is built on top of InfluxDB. The shift towards open source in manufacturing is driven by trends such as Industrial IT, where historians play a crucial role in data management and analysis. Open source adoption in manufacturing companies is becoming increasingly popular due to the availability of protocols like OPC-UA and InfluxDB, making it possible for smaller companies to compete with larger ones. Historians are ripe for disruption, thanks to open software, allowing them to adapt to changing needs without significant functional changes. Factry has helped large companies transition from incumbent historian vendors to its own solution based on InfluxDB.

In 2019, InfluxData predicts that containerization and Kubernetes will continue to gain popularity among DevOps teams, leading to changes in monitoring trends. The shift to container architectures is making legacy monitoring tools unsuitable for this new deployment architecture. As a result, there are expected trends such as the evolution of metrics and events observability, growth in CX Monitoring beyond infrastructure and toolchains, traction for time series platforms, an increased need for purpose-built platforms, and additional ground gained by Kubernetes while workloads still operate on non-K8s infrastructure.

Telegraf 1.9.3 has been released with several improvements to its plugins, including Agent, AMQP Consumer Input, Couchbase Input, Prometheus Input, and SQL Server Input. The updates address issues such as metric buffer ordering, credential leaking, and arithmetic overflow. These enhancements aim to improve the overall reliability and performance of Telegraf.

InfluxData has hired Jim Walsh, a former Salesforce executive, as its new Senior Vice President of Engineering, bringing over two decades of experience from Microsoft where he co-founded the Bing engineering team and started the Cosmos data platform. Walsh also co-founded cybersecurity firm Versive, which was acquired by eSentire in 2018. He expressed his excitement to join InfluxData, highlighting the company's leadership in time series platforms and its potential to derive value and insight from increasingly sensorized environments.

Here is a summary paragraph of the given text: GraphQL makes a powerful companion to your InfluxDB database, allowing clients to specify exactly what information they need without over-fetching or under-fetching data. Combined with Flux queries, this provides two loci of control for transforming and selecting data, empowering clients to make decisions about the data they receive. This setup creates flexibility in the backend while giving clients power over their data choices, making it easier to avoid superfluous code and adapt to changing requirements without altering endpoint URLs.

InfluxData, the creator of the modern open source time series database InfluxDB, has appointed Jim Walsh as its new Senior Vice President of Engineering and Will Paulus as its new Vice President of Sales. The appointments are aimed at propelling the company's plans for growth and expansion, building on its strong foundation in delivering innovative technology to DevOps and IoT applications. With Walsh and Paulus joining the executive team, InfluxData is well-positioned to innovate, expand, and increase profitability, leveraging their entrepreneurial initiative and extensive experience in scaling technology and driving sales growth. The company's time series database platform has gained significant traction across industries, including manufacturing, financial services, energy, and telecommunications, with over 450 customers including Cisco, eBay, IBM, and Siemens.

As SVP of Engineering at InfluxData, Jim Walsh joined the company to enable its co-founder and CTO Paul Dix to focus on his role as visionary while helping the engineering team grow, scale, and execute at an amazing velocity. Walsh is a builder with a strong track record of founding and leading successful companies, including Cosmos, which powered Microsoft's large-scale services, and Salesforce's data platforms organization. He sees immense value in data, which drives rapid scientific and technological improvement, and believes InfluxData is uniquely positioned to capitalize on the growing trend of instrumentation in both virtual and physical worlds. Walsh values integrity, impact, passion, and family, and is impressed by the team's high level of integrity, open source ethos, and commitment to remote work, which he believes will enable the company to attract and retain top talent from around the world.

I was inspired by the company's product, people, culture, and opportunity after meeting with Evan Kaplan, the CEO of InfluxData. The time series space is a perfect fit for my background in real-time analytics, particularly with their leading time series database technology, InfluxDB. I'm excited to join a world-class team at InfluxData, which has seen fantastic customer and revenue growth through its unique open source distribution strategy. The company's executive team, including Evan Kaplan, is technical, passionate, and collaborative, making it an ideal place for me to contribute my skills and experience.

Flux is a new language developed by InfluxData, used for querying and analyzing data in InfluxDB. Katy Farmer, an InfluxData Developer Advocate, provides a tutorial on how to window and aggregate data using Flux queries. She uses a script that routes cryptocurrency exchange information into InfluxDB and aims to compare the actual trading price with expectations. The tutorial demonstrates various steps of windowing and aggregation in Flux, allowing for easy comparisons between Flux and InfluxQL queries. By using Flux, Farmer can make meaningful comparisons and gain insights from her data.

In distributed systems, observability is a critical component for understanding system performance and identifying issues. Gianluca Arbezzano emphasizes the importance of distributed tracing as an essential tool in achieving observability. He draws on his experience at InfluxData to highlight the complexities of modern applications and the need for distributed tracing in development cycles. Observability, he notes, is not limited to production environments but also plays a crucial role in development, where it helps developers understand how their programs work and identify potential issues earlier. By adopting distributed tracing, organizations can gain valuable insights into their systems, improve performance, and reduce downtime.

New maintenance releases for InfluxDB OSS and Chronograf are now available, addressing issues with invalid UTF-8 bytes in database or measurement names and adding support for Flux query logging and user authentication. The updates include Flux 0.12.0, updated documentation, and fixes for data source connection editing issues.

In a recent podcast discussion with The New Stack, Paul Dix, Founder and CTO of InfluxData, discussed the growing need for time series databases, citing millions of downloads and enterprise customers. He also highlighted the importance of understanding data in a time dimension, suggesting that the time series space is just beginning to be explored. Additionally, Dix shared information about Flux, InfluxData's new query and scripting language for time series data, which is expected to play a key role in this growing field.

InfluxDB has released maintenance updates for both its Enterprise and OSS versions, with version 1.6.5 being made available. The updates are now live and can be downloaded from the InfluxDB website or through the InfluxDB Enterprise portal for Enterprise customers. For those who need help, support is also available to assist with any issues or questions.

StreamSets is partnering with InfluxData to integrate their time series technology into StreamSets' multi-cloud DataOps platform, aiming to deliver a more modern monitoring approach for cloud-native and multi-cloud projects. This partnership seeks to address the growing need for companies to modernize their monitoring solutions due to the increasing prevalence of newer app trends such as DevOps, AI/ML, and IoT. InfluxData's time series technology is expected to provide a solution that can offer real-time data insights at 99.99 percent uptime levels, unlike traditional monitoring tools which only provided information at one-second intervals.

David McKay recently joined InfluxData as a Developer Advocate. He has experience in various roles such as Software Developer, Tech Lead, System Administrator, and SRE. His interests include Alphabetized Lists, Cloud Native, Containers, Esoteric programming languages, Event Sourcing, Kubernetes, Microservices, among others. InfluxData's commitment to open-source and positive feedback from employees on Glassdoor were key factors in his decision to join the company. McKay found the interview process enjoyable and appreciated the company's transparency regarding its core values.

In 2019, InfluxData's open-source Time Series Platform, InfluxDB, has seen significant growth with millions of downloads and an expanding list of enterprise customers. The company's success validates the need for robust time series solutions, with Amazon's recent announcement of Timestream as a competitor in 2019. As competition fuels innovation, it is expected to drive growth in the time series category as a whole.

The author is learning Flux, a new language developed at InfluxData, to analyze cryptocurrency exchange data. They start with a basic query using the `from`, `range`, `filter`, and `window` functions to extract specific fields from the "crypto/autogen" bucket over the past 48 hours for Bitcoin (USDT_BTC). The results are overwhelming, so they filter out unnecessary data and focus on just Bitcoin. They then apply a window function to group the data by hour and calculate the mean of each field. Finally, they use the `group` function to combine the relevant values into fewer tables. The author finds the Flux query readable and easy to understand, with only one dislike - understanding the table structure being returned. Overall, they successfully write a useful Flux query that allows them to make meaningful comparisons in their data.

A new maintenance release of Telegraf 1.9.2 has been made available, addressing various issues and improving several plugins including Agent, AMQP Consumer Input, Azure Monitor Output, Ceph Storage Input, Docker Input, Network Response Input, Parser Processor, Prometheus Input, SQL Server Input, Strings Processor, Syslog Input, Varnish Input, and Windows Performance Counters Input. Notable fixes include resolving issues with global metric modifications, empty message consumption, tag value sending errors, and counter processing for certain inputs. The release also includes updates to plugins such as Docker Input, which no longer panics when the URL is not properly configured, and Syslog Input, which now supports non-transparent framing of syslog messages.

InfluxData has released maintenance updates for both InfluxDB Enterprise and OSS. These updates are not the latest versions but provide fixes to support community members who do not want to upgrade immediately. The current version of InfluxDB OSS is 1.7, with a forthcoming release of 1.5.5 available for community members.

The master branch of InfluxDB will be switched to the alpha version of 2.0 on January 11th, marking a breaking change from 1.x. The current 2.0 work is happening in the platform repo, which will eventually replace the archived 1.x codebase. From then on, all commits to master will focus on the 2.0 line, with quick iterations and versioned alpha releases planned for the coming weeks. This phase aims to gather community feedback and iterate on features before moving into beta mode and eventually releasing the final 2.0 version.

InfluxData recently concluded a series of six "office-hour" webinars on Flux, its new functional data scripting language designed for querying, analyzing, and interacting with data. The goal was to answer Flux-related questions and gather community feedback to improve the language. During these sessions, InfluxData's Flux engineer Adam Anthony showcased cool features in the products, did amazing demos, and even replicated a user's InfluxQL-based query and re-wrote it in Flux language. These sessions demonstrated the versatility of the language and its capabilities for handling time series data and beyond. The webinars covered various topics, including aggregateWindow, percentile functions, group function changes, and helper functions, providing users with a comprehensive understanding of Flux.