Deep Dive into InfluxDB 3.0 – An InfluxData Partner Update

Session date: Dec 05, 2023 08:00am (Pacific Time)

See you on December 5, 2023, for an exclusive webinar tailored to our valued InfluxData partners. We are excited to provide an InfluxDB product update and present collaboration best practices. As the purpose-built time series database, we rebuilt InfluxDB’s core in Rust and it sits on top of Apache Arrow and DataFusion. InfluxDB 3.0 offers a new columnar storage engine that uses the Apache Arrow format for representing data and moving data to and from Parquet. During this webinar, you’ll learn the latest InfluxDB releases benefit from these technologies and why we picked Apache Parquet as the persistent format.

In this webinar, Gary Fowler and Michele Todd will highlight:

• The ins and outs of InfluxDB 3.0: InfluxDB Cloud Serverless, InfluxDB Cloud Dedicated, and InfluxDB Clustered
• The vision behind our latest releases and how they can benefit your clients and projects
• Our InfluxData Partner program update and a sneak-peak of what’s coming in 2024

Don’t miss this opportunity to connect with our team to discover how the latest InfluxDB innovations amplify your product and service offerings. We’ll round things out with some important partner updates and a live Q&A session.

Click here for presentation

Watch the Webinar

Watch the webinar “Deep Dive into InfluxDB 3.0 – An InfluxData Partner Update” by filling out the form and clicking on the Watch Webinar button on the right. This will open the recording.

[et_pb_toggle _builder_version=”3.17.6” title=”Transcript” title_font_size=”26” border_width_all=”0px” border_width_bottom=”1px” module_class=”transcript-toggle” closed_toggle_background_color=”rgba(255,255,255,0)”]

Here is an unedited transcript of the webinar “Deep Dive into InfluxDB 3.0 – An InfluxData Partner Update.” This is provided for those who prefer to read than watch the webinar. Please note that the transcript is raw. We apologize for any transcribing errors.
Speakers:

• Caitlin Croft: Director of Marketing, InfluxData
• Gary Fowler: VP of Products, InfluxData
• Michelle Todd: VP of Business Development & Partnerships, InfluxData

CAITLIN CROFT: 00:00
Once again, welcome to today’s webinar. My name is Caitlin Croft. I’m joined today by Gary Fowler and Michele Todd, who will be providing a deep dive into InfluxDB 3.0 and provide a really fantastic update on the InfluxData Partner Program. Please post any questions you may have for them using the Q&A. This session is being recorded. And without further ado, I’m going to hand things off to Gary and Michele.

GARY FOWLER: 00:30
All right. Thanks, Caitlin, and thanks, everyone, for joining our partner update and deep dive into InfluxDB 3.0. My name is Gary Fowler. I’ve been with Influx about a couple of years, but this is actually my second day installed as the VP of Products at InfluxData. So, a recent change for us. I am going to start off the presentation, and then Michele Todd, who is our VP of BD and Partnerships, is going to jump in, talk about our partner program. So, a fairly full agenda here. We’re going to do a quick introduction on InfluxData in case there’s anyone that doesn’t know about InfluxData already. We’re going to talk about some common use cases that we’re used for. We’ll talk about our product suite, and then we’ll spend most of our time in sections four and five talking about the InfluxDB 3 and how it relates to our partners and what opportunities it creates for our partners, and then we’ll have some time at the end for Q&A. All right. So, I think everyone here probably has heard of InfluxDB or knows about InfluxDB, and that is why you’re joining. If any of you has not, we are a time series database. We’re built for time series data and real-time analytics, built for high-scale availability and a lot of management options. We have a very big adoption out there in the industry. We’re the runaway market leader in the time series category. We have almost a million active instances of InfluxDB out there. Our cloud implementation, we have over 100,000 accounts with it. And in the open-source community, GitHub, we’re almost 27,000 GitHub Stars.

GARY FOWLER: 02:19
There’s certainly a lot of use cases that people use InfluxDB for. Here’s some of the most common ones: infrastructure, network, and DevOps monitoring, they’re all popular. IoT data is certainly also in our sweet spot. We have a lot of customers that use us for IoT sensors and things like that. And really, any kind of analytics SaaS applications is very often a good fit for us with business intelligence and machine learning input and different things like that. What you may or may not know is you’ve probably all used InfluxDB. You just may not have realized it. So, here’s just a few examples. Here’s a few of the products that use InfluxDB. Tesla Powerwall isn’t just an application that you have on your phone to monitor what you’re doing. It has many components. There’s solar panels, battery storage, all reporting data that gets reported in the mobile application, and all that data ends up flowing through InfluxDB. If you have a Nest Thermostat, you’ve likely used InfluxDB. It’s another company that’s known for its innovation. It uses InfluxDB to monitor their entire Nest infrastructure. If you stream Disney movies or programming, you’ve probably used us as well. Disney+ is a video streaming pioneer that uses us to track user behavior and other custom metrics. And Rappi is one of the largest online e-commerce marketplaces in South America. So, if you’ve been down there and you’ve used that delivery service, you’ve also used InfluxDB.

GARY FOWLER: 03:59
So, here’s a few examples of customers who’ve adopted Influx. One of the things I’ve loved since I joined Influx a couple of years ago is how many great companies are building really fun and innovative things on our platform. So, I’m a football fan, so it’s always fun for me to look at a slide like this and see NFL.com. And if you’re a gamer, it’s fun to look on here and see EA and some of the other things on here. And certainly, I think we’re all proud of the customers that we have in the sustainability and clean energy space up here that you see in these logos. All right, so just a quick word about our previous existing product suite. We’re mainly going to focus on our latest InfluxDB 3 versions. But those of you that have been working with us or have worked with us in the past, you probably know us from some of our previous products, including our InfluxDB open source software, our Enterprise software, which is our high availability self-managed software, and then our cloud implementations. We have both a single-tenant pay-as-you-go solution or a multi-tenant pay-as-you-go solution and a dedicated single-tenant version of our cloud software. But what you may not be familiar with is our InfluxDB 3.0 suite, as we are just rolling it out this year. This includes our multitenant cloud solution, which we call InfluxDB Cloud Serverless. It has free and pay-as-you-go options. Our InfluxDB Cloud Dedicated platform is for our enterprise customers. And in an early preview, our InfluxDB Clustered product, which is a self-managed version that you can use on-prem or in private cloud. So, it’s those implementations that we’re going to be talking about today. We are working on an InfluxDB 3.0 open-source version and a freemium version. So, we’re hoping to come out with those later, towards the end of 2024, maybe slightly into 2025, but we are still advancing our open-source versions of the software as well.

GARY FOWLER: 06:13
All right. So, let’s get right into why you probably came here today, and that is to hear about InfluxDB 3.0. So thought I would start by just explaining why we built it. Why did we build this thing? So first, we were really trying to respond to our customer needs and demands. While some customers really liked our proprietary but powerful query and scripting language, Flux, there were many that really wanted to use something they were more familiar with and wanted us to support SQL. So, others preferred that we bring forward our InfluxQL software, so one of the things that we wanted to do in InfluxDB 3 is to offer SQL support. Some of our customers with very large data sets were really pushing the cardinality limits of our previous versions. We, of course, wanted to give them additional options, but the architecture of our previous version made it difficult to continue to raise those cardinality limits, and so we built InfluxDB 3 for unlimited cardinality. Customers also wanted to be able to use third-party visualization and other tools beyond what was already supported in previous versions. We had to write connectors and things for any third-party software that we wanted to use. So, we wanted to build something that used a little bit more open standards and allowed for more third-party interoperability. And then we wanted to reduce storage costs. We wanted to make InfluxDB more efficient, reduce the storage costs for ourselves, for our cloud software, in turn allowing us to offer more aggressive pricing, but also to make it more economical for customers that were running it in their own private cloud or on premise.

GARY FOWLER: 07:59
Another big thing that we saw besides these specific requests for customers was really the evolution of our customer and their workload. So initially, with time series data, a lot of it was being used purely for observability and alerting, right? You might have a sensor out there, you want to set a threshold, say, “Hey, if the sensor reports a value higher than this, that needs to set off an alarm somewhere, some sort of action, we need to do something with it.” Or we had something that we would call a dead man check that we would do in InfluxDB Cloud 2, for instance. And the way it would work is if there was a sensor that didn’t report for a period of time, that would set off an alarm. And those types of things were a lot of what people wanted to use time series data for. But we certainly, over the last few years, started to see a lot more of our customers use it for real-time analytics to do trend analysis, right, anomaly detection, different things like that. So, we wanted to build something that was going to be even better than what we had before for those real-time analytics.

GARY FOWLER: 09:09
Okay, so those were some of our goals and objectives when we set out to build 3.0. It wasn’t an easy project. It took our engineers two and a half years or so to do it. And so how did we end up doing it? Well, first, we started by building out a tech stack that’s been gaining popularity and traction, especially in the data science community. Apache Arrow is an in-memory columnar data store. Apache Parquet is its on-disk equivalent. Many developers already work with both Arrow and Parquet as a columnar database format. The Apache Flight SQL and DataFusion help give us the SQL interface to both then in-memory and on-disk storage. So, we started with the Apache ecosystem, and then we built on top of that. So, we added several key components because just having Parquet or Arrow alone doesn’t necessarily scale that well if you want high-performance queries, fast queries, and ingestion, so we built a number of things on top of it. We had a high-scale ingestion with zero TTBR, that’s time to be ready, as queries will check with ingestor nodes for the latest hot data. So, we built these ingestor nodes. When you do a query, the ingestor—or when you insert data into the database the ingestor receives it, but it also makes that data available in hot memory for instant query ability. We added a catalog service for quick metadata retrieval and data location. So, things end up getting stored in multiple Parquet files. The catalog helps determine, based on the query that you’re doing, which Parquet files to look in to extract the data.

GARY FOWLER: 10:53
We added a compactor service which takes multiple Parquet files and consolidates them into logical partitions of Parquet files. And we added support for low-cost object store for the cold data. We started with the Apache ecosystem. We built a whole bunch of components on it. What we ended up with in our design was we wanted something that was designed for both hot and cold storage, because we wanted to deliver both performance and low cost. So, we made the most recent data that you write be in hot memory, right? It’s hot. It’s data that’s in memory. It doesn’t have to go to disk to get it. So, when you’re doing queries based on the data that was just loaded, observability analytics, otherwise, we go right to memory, and we can retrieve that quickly. For things that are older, weeks, months, years older, we put that in object storage. So, it’s a low-cost place to store the data, and we keep the cost low that way. So, this is kind of what the high-level architecture ended up looking like when we were done. You can see things in this diagram that are based on the Apache Arrow ecosystem has a little feather by them, but you can see some of the other components that we built in as well.

GARY FOWLER: 12:15
So, what’s been our result so far after building this thing? So, a couple of things that we noticed right away as we started rolling this thing out. One, this hot data in memory has allowed us to do real-time, right? The second data is written, you’re able to query it, get a full picture of what has been written, subsecond responses for recent data. That’s went really well. The low-cost storage, I think we were even surprised to see how much the columnar database allows us to compress the data and use a lot less storage. In addition, putting it in an object store makes it cheaper storage, so it’s a lot less storage and cheaper. And then we don’t have the cardinality constraints that we had with previous versions. So that’s been great. So fast, real-time, and cost-effective. We’re seeing 45 times the performance with our write performance. So, this is the ingest data coming in. Between the compression making less data and the object store making it a lower cost data, we’re seeing about a 90% reduction in storage cost. For high cardinality data, we’re seeing 100 times faster queries, we’re seeing 45 times faster queries for other recent data. This slide, if you can see, it’s kind of a busy slide, but shows some of the numbers comparing InfluxDB 3 with InfluxDB OSS version, and you can see the difference. On the upper left, you can see the improved ingestion performance. You can see the numbers there. Upper right, you can see some numbers for our query improvement. And in the bottom, you can see that disk compression benefit that we’re getting.

GARY FOWLER: 14:11
The other thing that we are seeing is improved interoperability that we get from using the Apache Arrow ecosystem. And when I say that we participate with the Apache Arrow ecosystem, I mean, yes, we utilize this in our products, but we’re also key contributors to this open-source community. So, we do a lot upstream with DataFusion. A couple of our engineers are regularly contributing to DataFusion and other Apache Arrow ecosystem projects. So, this participation, one of the things that allows us to do is more easily integrate with well-known platforms and library selections such as Plotly, the Anomaly Detection Toolkit, Apache Spark, Tableau, Grafana, RapidMiner. So, we’re just seeing this list grow. As more and more customers—or more and more vendors also work with Apache Arrow, you’ll see this list continue to grow. Some of the early buzz from our customers, they love the near-zero time to be ready. In previous versions, like in our cloud software, for instance, the ingestion ran through a Kafka queue, which was pretty quick, but there was still a couple of second delay in the time that you wrote the data to the time that it was available for query. So, they love that it’s available right away. Those customers that have already been using Arrow and Parquet certainly love it and like the third-party integration that it brings us.

GARY FOWLER: 15:52
The SQL support has been very popular. Our InfluxQL diehards are glad it’s become a first-class citizen. Those of you that have been working with InfluxDB for a while, you know that our early versions, we had InfluxQL as our query language, and then we kind of let it sit when we started developing Flux. But now we’ve brought it back. We’ve rewritten it in Rust for improved performance and interfaced it with DataFusion so that you can use InfluxQL with InfluxDB 3. And we also have customers that really like our new v3 lightweight client library, which allows you, from the popular languages, to ingest data or query data. And when you query data, you can get the data back in an Arrow format, which our data scientists that work with things like Pandas really like. So, in summary, we think InfluxDB 3 is fast and efficient. We think it provides a low-cost storage, resulting in a lower overall cost of ownership, even in comparison with previous open-source versions. We’ve again improved our time to awesome with SQL and improved third-party interoperability. It’s designed and built to deliver subsecond query responses, especially for recent data. This allows you to build awesome end-user experiences. Also, with InfluxDB 3, you don’t have to choose between storing data forever and keeping your costs low. We’re also very developer-focused, whether it’s SQL native support or support for InfluxQL, which is SQL-like but much better in terms of working with time series data. It’s also designed to be interoperable with other machine learning tools or data lake with zero copy sharing and other features that improve data efficiency by a factor of two or three. And with that, I am going to turn it over to Michele Todd to talk specifically about our partner program and what all this means to you.

MICHELE TODD: 18:00
Okay. Just going to share my screen here. Okay. Thank you, Gary, and hello, everyone. My name is Michele Todd, and I head up partnerships here at InfluxData. I just came back from AWS re:Invent in Vegas last week, and I had so many great conversations with our partners, which makes me super energized about what we can do together. As Gary mentioned, we help customers, which makes me super—which helps us achieve customers having time to awesome. That basically means we help them build cool stuff and solve problems quickly. And all of this would not be possible without the invaluable support of our partners. I joined just a few months ago, and my goal coming in is to really reinvigorate our partnerships and take them to the next level. InfluxData plays a pivotal role in a multitude of scenarios, enabling the delivery of enriched user experiences built for time series data. Whether you’re a technology partner, a reseller, or a distributor, or a consultancy partner, we really want to empower you in better serving our mutual customers. So, first of all, let’s talk about our technology partners. Technology partners play a pivotal role because InfluxData is an infrastructure product never used in isolation. So, you all build the technical integrations that are crucial for making InfluxDB operate efficiently across time series use cases.

MICHELE TODD: 19:29
InfluxData offers open APIs, facilitating seamless integration with various technologies, just like Gary mentioned, allowing for developing scalable solutions on our platform. We really urge you to join our vibrant open-source community to collaborate, to innovate, and to contribute to the evolution of our products. If you build an integration with InfluxDB, you can really count on our support, too. Our team will review your integration, offer you guidance, and ensure it meets the highest standards of performance. You can showcase your integration proudly on your website with our built-on InfluxDB badge as well. Some of our technology partners go beyond integration, incorporating InfluxData products into their proprietary commercial offerings. Through our OEM agreement, we grant partners the rights to seamlessly embed InfluxData technology within their solutions. This not only ensures a smooth integration process but also empowers partners to deliver enhanced value by leveraging InfluxData’s purpose-built time series platform. One great example that I love is our collaboration with PTC and their ThingWorx IoT cloud platform. Partnering with us extends beyond technology. We’re dedicated to showcasing your integrations through joint marketing efforts such as blogs, webinars, case studies, and featuring you on our website. This will really boost your solution awareness and drives demand for both of us.

MICHELE TODD: 20:57
Lastly, we can collaborate with your mutual partners to drive the adoption of your solution, whether it’s a system integrator or even a cloud provider like AWS that we work with heavily. It will really create a win, win, win. So let me highlight a compelling example of a technology partner that we work with called Terega. So Terega faced challenges with their aging on-premise IT system and the need to onboard customers swiftly. To address this, they modernized their infrastructure with InfluxDB as the foundation, creating their cloud-native historian IO-Base. By working closely with our team, they chose InfluxDB Cloud, and they overcame scalability issues, enabling faster onboarding of customers. Their proprietary connector, IO-Base, seamlessly ingests data from on-premise gas infrastructures in real time to InfluxDB Cloud. Leveraging our APIs, they integrated advanced analytics, providing real-time insights to field technicians and partners. So, what was the result? They now collect and store higher granularity data 10 times more than their legacy solution, reducing their overall TCO by 50%. Really, really cool use case. So, we recently did a joint webinar with them as well as a case study, and we actually just featured them in our re:Invent booth last week. So, we’re building a great pipeline, and we have some active joint sales opportunities going with them. So just wanted to highlight that this is the type of activity that we would love to do with more technology partners.

MICHELE TODD: 22:29
Our global network of resellers and distributors also play a vital role. They foster vendor relationships with our prospects and customers who prefer transacting through them rather than directly with us. This ensures a seamless experience for our diverse clientele. Our partner coverage spans companies dedicated to providing solutions across all vertical markets we cater to, such as manufacturing, industrial IoT, energy, financial, and more, as well as geographically across North America, Europe, and Asia Pacific, which ensures a broad reach. We really do see them as an extension of our salesforce. We recently enhanced our agreement, so that’s super exciting. This enables the resell of our cloud solutions. Previously, our partners were only able to resell our on-prem solutions. That’s a great change that we made this year. Additionally, we’ve recently just launched a new deal registration process. By registering deals, partners can gain access to enhanced deal support and dedicated resources and really making sure that we give you a streamlined and efficient sales process throughout the sale. Today, I’m also excited to introduce new badges that are represented below on the slide here. And I think this is a great way for you to—just a visual representation of our partnership, and you can proudly display it on your websites. It’s a great way to showcase your commitment and collaboration with us.

MICHELE TODD: 23:54
And as we are onboarding more and more resellers and distributors, we’re really looking for partners that can provide additional value-added services. For example, we have many partners that provide tier-one support, so just as an example. And then I also want to mention coming in 2024, we will be launching our AWS Marketplace CPPO coverage, which is their Channel Partner Private Offers. This will allow partners that can resell our software on our behalf, and it allows customers to tap into the advantages of AWS Marketplace, such as their EDP Cloud. So, stay tuned for more on that next year. And then lastly, our consultancy partners and our systems integrators, they play a pivotal role in weaving together tailored solutions to address unique customer needs. Armed with industry best practices and a pulse on the latest trends, consultancy partners ensure enterprises benefit from the most current and effective solutions. So, by choosing InfluxData, you’re really endorsing a best-in-class time series solution that’s flexible, scalable, and precisely tailored to your specific customer’s use case, and it can bolster your RFP as the strongest available option. Unlike some ISVs, we intentionally avoid offering our own professional services to prioritize our partners’ success.

MICHELE TODD: 25:17
Our primary focus is to build the best time series database, since we’re a software company. And we believe that to create a thriving ecosystem, it’s essential to have a strong network of trained partners proficient in InfluxDB. This ensures that when we refer our customers to you, we do so with confidence, knowing they are in capable hands and that our combined efforts will lead to successful projects. So, to empower our partners with the right skills and knowledge, we offer top-notch enablement and training. Options include our weekly webinars. I think Caitlin mentioned some of these weekly webinars or InfluxDB University online courses, and we will be releasing some new partner deep dive training in 2024. You can also sign up for our free cloud trial up on our website. But based on customer—or partners’ request, we have been asked to provide not-for-resale licenses for up to six months to our partners. This allows you to explore and to experiment with our technology for DevTest and training purposes. Once trained and equipped, we facilitate tech reviews with our SE team upon identifying a customer use case or integrating Influx into your reference architectures. And then from a go-to-market perspective, we’re here to help you. So, we have marketing development funds that can be used with some of our top partners that we are collaborating on joint opportunities with that will continue to build that shared pipeline. And lastly, I also wanted to mention our referral program. This is designed for partners that may not want to necessarily resell products, but it gives additional incentives for you for your valuable efforts. So, if you are interested in that, definitely let us know.

MICHELE TODD: 27:05
So, I wanted to just pause for a moment and just touch on something that Gary touched on earlier because I think this is something that really, really resonates with partners we talk to, especially the ones we talked to at re:Invent last week. We often see our partners start out by working with our open-source software—or working with customers that are working with our open-source software. So based on the latest 3.0 products that Gary just walked through today, you heard about the many benefits over our previous versions. However, one thing that we really want you to walk away with from today is knowing that your customers could actually see an overall reduction in TCO based on the storage savings alone. The two main factors contributing to this TCO reduction are 3.0 storage compression, which is four to five times better than previous InfluxDB versions, and cost of object stores such, which is three to four times cheaper than SSD-based storage. Together, they enable customers migrating to 3.0 to have at least 90% reduction in storage costs and therefore, better TCO. I think it’s important to keep in mind as you are working with your customers that have traditionally used our open source and been happy and they think it’s just good enough that actually they need your support when they’re trying to make that decision. Because we oftentimes talk to customers that are like, “Well, I’ve been using open source. Why should I switch to 3.0? What’s the benefit there?” And I think this is the strongest story and use case that we need to really make sure that our customers are aware of. And then last of all, before we wrap up, I wanted to make sure that we covered off on just these integration service opportunities that are exciting with 3.0. It’s not just about the features but how these features open doors for collaboration and innovation for all of our partners like you. So maybe Gary, do you want to jump in and just touch on some of these excellent and exciting opportunities?

GARY FOWLER: 29:02
Yeah. Absolutely. So, we do think there’s a lot of opportunities for any of our partners that are doing system integration based on a number of factors. So, first is the third-party integration that we’ve talked about a few times with the things that Arrow Parquet brings up. So, one of the things that is available for Apache Arrow and Parquet with DataFusion is a Flight SQL JDBC driver. And what that allows is really any application that can use a Flight SQL JDBC driver—any application with JDBC to be able to use that Flight SQL JDBC driver for interoperability. So that means if you’re doing—Tableau, for instance, is one that our customers are starting to use quite commonly because of this. So, if you’re also doing work where you’re doing Tableau consulting or you’re helping a customer with that, it makes it easy to set that up as an additional data source. The query language, one opportunity we’ve created for our systems integrators, is that we’re not carrying Flux forward into InfluxDB 3, at least all flavors. And so that creates some conversion and consulting opportunities for our system Integrators to help customers convert what they were doing with Flux into SQL or InfluxQL.

GARY FOWLER: 30:30
Downsampling, we have not yet built in downsampling into InfluxDB 3.0. We do plan on doing it at some point in the future, but we’re not there yet. But we have some examples on containers that can be taken or do-it-yourself or partner-assisted downsampling. It’s configuration-driven, so you don’t have to write it from scratch. Just take parameters like, hey, what kind of aggregation you want to do, what are the fields, etc. And you can configure it and deliver downsampling capability to customers. Our new v3 library allows customers to work with data in that Apache Arrow columnar format. So, some of our customers or some of your customers might need help getting started with the v3 client library if they’ve used our v1 and v2. For those that have data scientists on board that’s using tools like Pandas, they’re going to love it if you can help them get started with that. And then InfluxDB 3 has different schema optimizations than we had in the past. So, customers that are upgrading from our open source versions, our Cloud 1 or Cloud 2 solutions or Enterprise, they may need to make some schema modifications, and that is something that you could consult with them on and help them with. And so, we do think that there’s a lot of different opportunities for our partners to step in and help build a complete solution for their customers using third-party applications. InfluxDB, of course, being a big part of that, but using third-party applications and integrating them all together.

MICHELE TODD: 32:20
Awesome. Thanks, Gary. Okay, so hopefully, the product update and the partner program update that we went through today provide you with some insights into our journey and where we’re headed. I think this hopefully can help you identify integration opportunities or other ways we can partner together. But before we jump into Q&A, I just want to walk through a few calls to action. So, first of all, get trained up. So, we actually have InfluxDB University, which has several courses. There’s a great 3.0 Essentials course if you haven’t taken that one. There’s a couple of popular Telegraf courses as well. We will be launching more InfluxDB courses throughout next year. And then also we will be announcing InfluxDays, which will be coming back in 2024, so stay tuned in the coming months on the dates, and hopefully, you can join us for that. We’d love to see you attend that. If you’re not already a partner, go to influxdata.com/partners-signup and fill out a form, and we will review your application and follow up on next steps. But we would love for you to join our partner program. And then the other big thing which I think has been covered in the day today, and Caitlin touched on it, is really actively participate in our community. Influxdata.com/community is our open-source community Slack forum where you can collaborate, ask questions, and become a subject matter expert where you can actually answer questions and also become an InfluxAce, which is a highly sought-after area as well. And then also we do have a partner Slack channel, so if you’ve become a partner and you either lost the Slack channel or you need access to it, let us know because that’s a way you can collaborate with other partners. Next, sign up for our free cloud trial. It’s a quick and easy way for you to get your hands on our software and check it out, try it out, and get even more trained up on it and understand the latest version.

MICHELE TODD: 34:22
And then finally, anything you need assistance on, please, please email [email protected]. So, whether this is you need to get in touch to do a tech review, you want to do a blog post with us, anything that we can do to generate more excitement and awareness about our joint solutions, I’m very excited to work with you, so please, please email [email protected] if you need any assistance. So, we will be doing these webinars every quarter to keep you updated. I’m really excited to collaborate with you all. I think together, we’re going to be able to really enhance the value we bring to our mutual customers and drive some joint revenue, as well as just create an ecosystem where everyone benefits. So super energized, super excited. Like I mentioned, I just joined a few months ago, and I’m just getting started and getting ramped up, but want to learn from all of you on what you’ve been doing and what we can do more to take it to the next level. So, with that, I’m going to go ahead and hand it back over to Caitlin to open it up for any questions.

CAITLIN CROFT: 35:25
Awesome. Thank you so much, Michele and Gary, great job. So, there is a question here. It says, “Hi. We are considering running the open-source version of InfluxDB on the Edge and replicate that data to the cloud-hosted instance. In both, we would like to run Tasks to analyze data streams in real time. My question is, can we, for now, run InfluxDB version 2 at the Edge and replicate the data from it to InfluxDB 3.0 in the cloud? Can we migrate then easily to InfluxDB Edge when it is released? If we invest in Tasks written in InfluxQL or Flux, will InfluxDB Edge support them, too? Or maybe we could stay with InfluxDB 2 forever. In that case, how long will it be supported?”

GARY FOWLER: 36:18
Yeah. So, there’s a lot of questions in there. And so let me address the first one because—let me address the Tasks one first because it’ll affect all of the other answers. We have not carried forward Tasks into InfluxDB 3.0 yet. So, Task was built on Flux, which, as I talked about when I was doing my section, some customers really liked it. But a lot of our customers, a larger set, really wanted us to focus on SQL and things that they already knew how to use or was a little bit more common in the industry and less proprietary to InfluxDB. So, we do plan on having a Task type replacement. We’re not there yet. We’re thinking of more of a virtual machine that can run on the platform that you could run Python or something like that instead of Flux to be able to do those things. But that’s not in place yet, so you would not yet be able to do what you’re asking to do. So, you can stay within InfluxDB 2. We’re not [inaudible] it anytime soon. We think it’s going to be there for a while. So no, at least, short-term worries about that within the next two or three years.

CAITLIN CROFT: 37:42
Awesome. Thank you. Will there be a migration path from InfluxDB 1.x Enterprise to InfluxDB 3.0?

GARY FOWLER: 37:53
Yeah. So, we released InfluxDB 3 in a couple of implementations already. It was really based on customer demand and wanting to get it out there quickly. Our migration tooling for earlier version to InfluxDB 3 is in the works, but it’s not completed yet. But hopefully, later in the migration from 2.x, we’re hoping to be available in the first half of 2024. For 1.x, there won’t be quite as much migration from 1.x, and the workloads will largely work with InfluxDB 3. The actual data migration, we’ll be working on tooling for that as well.

CAITLIN CROFT: 38:45
Cool.

MICHELE TODD: 38:46
And the one thing I’ll just add to that really quick is I think that this could actually really open up some opportunities for our SIs while we are kind of getting our migration tooling out is what can you do from a services perspective to get in and help customers that are kind of making that jump and you might be able to help them in that journey until we have the migration tooling all available.

CAITLIN CROFT: 39:12
And Michele, you’ve been here for a while. I feel like you’ve been here longer than you have been. In 2024, what really excites you about all of our different partners, and what are you really excited for?

MICHELE TODD: 39:27
Yeah. I think that as I’ve been ramping up, I just realized that we just have such a big ecosystem. I mean, we are a developer company, right? So, I think that with that we have so many technology partners and so many SIs and consultancy partners that use our products. It’s how do we capture that and really start collaborating together? Because I think that oftentimes you’re off in the Wild, Wild West working with our customers, and we don’t even know about it. So, it’s how do we collaborate, connect, network? I want to do more. I’m going to try to do some in-person partner meetups next year, so I’m hoping to get over to Europe in kind of the—late February, early March is the goal to get over there, do something in North America after that. Like I said, the InfluxDays is coming back, so would love to kind of connect there. But I think that just getting more engagement to really understand how we can help our partners, because I think that how we can really work together is going to help a ton. The other thing too, is our latest 3.0 products are currently only available on AWS. It will be available on Azure and GCP later in 2024. But I’ve been working really closely with AWS, and I worked a lot with them in my past lives with Splunk and PagerDuty and Chef. So, I think that AWS has such a huge amount of partner ecosystem tools and similar to what I talked about earlier with CPPO, I just think there’s things that we can tap into with AWS’s help to just round it all out. It’s like the power of three. So, I’m really excited to see how we can work closely with AWS with you all to just take it to the next level.

CAITLIN CROFT: 41:24
Absolutely. And I think that’s what’s really cool about InfluxDB as a product, especially from a partnership standpoint where we have partners who are using it, as well as helping implement it, so it’s kind of, “We are helping you from both sides.” So, I think that’s pretty exciting. Let’s see. Someone’s asking, “Can we explore InfluxDB 3.0 Clustered? Can we obtain test licenses for it?”

MICHELE TODD: 41:56
So, I’m not sure if Gary, you want to answer that. As far as I know, it’s in private preview, so we can’t quite get it into the hands of our partners just yet. But I think in - what? - Q1, Gary, is what we’re hoping.

GARY FOWLER: 42:10
Yeah. We’re hoping that it’ll become generally available towards the end of Q1. Don’t know that for sure yet, but that’s what our target is right now.

CAITLIN CROFT: 42:22
Cool. And everyone on this webinar at the very least should have my email address. So, I’m always happy to connect you with Gary and Michele if you guys have further follow-up questions. What is the performance difference—what was the performance difference when retrieving cold data in InfluxDB 3.0 versus InfluxDB 1, particularly for an aggregation spanning several months?

GARY FOWLER: 42:49
Yeah. Very good question. And query performance for cold data is something that we continue to work on in InfluxDB 3. So right now, for that type of query, I would say that you probably have better performance with version 1 for long-spanning queries, queries that go across a long time period.

CAITLIN CROFT: 43:17
Can you please clarify: can we replicate today’s data from InfluxDB 2.0 open source running at the Edge to InfluxDB 3 running in the cloud?

GARY FOWLER: 43:30
I believe you can, but I don’t know the answer to that for sure. So, we can research that and get back to you on that one.

CAITLIN CROFT: 43:41
Perfect. And Gary, I’m just curious. You’ve been here for a while, you’re pretty familiar with InfluxData. Kind of twofold, what’s some tips and tricks that you just wish that everyone knew, something that you see people ask about that you just wish was out there in the broader ecosystem? And then my second question is, what excites you about the product roadmap, especially with 3.0 in the coming year?

GARY FOWLER: 44:10
I’ll answer the last question first. There’s a lot to be excited about, right? So, when you’re in product management, one of the things you’re doing is trying to define MVPs of, “Okay. What do I need to get this product to the market and get it into the hands of the customers that are waiting for it?” But of course, there’s a whole bunch more things that you want to add on top of that, right? So, we’re working on a number of things, including query performance, so manageability tools. We do want to have the migration tooling that we talked about. We do want to have a Task type replacement, a way that you can do tasks. So, we want to integrate some machine learning and some AI into the platform. There’s a lot of things that we plan to do over the next year or two that we’re excited about. Tips and tricks that you asked about: I think the biggest tip I can think of for anyone starting to use any of the versions of InfluxDB 3 is even though we are a schema on write product, meaning that you don’t necessarily have to predefine a schema, still important to have an idea of what you want your schema to look like because you can optimize it for query performance, you can optimize it for storage. There’s a number of ways that you can look at it based on how many columns you want to have, what you end up using as measurements and tags. So, I would say that is the biggest one is an opportunity for partners to provide that consulting to do that kind of schema, help the customers with their schema design.

CAITLIN CROFT: 45:55
Awesome. Thank you so much, and thank you, everyone, for joining today’s webinar. I will just stay on here just another minute or so, see if you guys have any other last-minute questions. In the meantime, Michele, Gary, are there any other last bits of wisdom that you’d like to share with the audience?

MICHELE TODD: 46:17
[crosstalk]—

GARY FOWLER: 46:17
I can’t think of last bits of—

MICHELE TODD: 46:20
Oh.

GARY FOWLER: 46:21
The only thing I was going to say is thanks for joining us on this partner update. Something that Michele would like to do regularly, quarterly. So excited to see you on this one and look forward to seeing you on future ones as well.

CAITLIN CROFT: 46:39
Awesome. Thank you.

MICHELE TODD: 46:39
[crosstalk]—

CAITLIN CROFT: 46:41
Michele, anything else you wanted to add?

MICHELE TODD: 46:43
No. I don’t think so. I mean, I think these are something that have not happened for a little while, so super excited to resurrect them. And yeah, just please, again, [email protected] is the best way to get a hold of me and my team. And yeah, just super excited to work with you all.

CAITLIN CROFT: 47:04
Awesome. Well, thank you, everyone, for joining today’s webinar. Once again, it has been recorded, and so it’ll be made available by tomorrow morning. So, if you want to share it out or you had to drop off, don’t worry, it’s being recorded and will be made available. So really appreciate everyone joining and thank you so much to Michele and Gary. You guys did an awesome job.

MICHELE TODD: 47:29
Thanks, Caitlin.

CAITLIN CROFT: 47:30
Thank you.

GARY FOWLER: 47:31
Thanks, everyone.

MICHELE TODD: 47:32
Thanks, everyone.

[/et_pb_toggle]