Migrating observability tools is easier than you think

Are you sticking with an observability tool that you don’t love? Plenty of organizations are unhappy with their current solution — whether it’s because it costs too much or because it can’t help troubleshoot issues quickly. Yet, they are reluctant to migrate to a new and better tool over concerns about hassle, time, uncertainty, you name it. 

But, did you know that migrating can take as little as two months, and you may end up moving over a mere fraction of your current assets? To back that up, we’ve asked four engineers to share real-world migration success stories from customers, to show why migrating tools is actually easier than you think.  

In a recent Chronologues episode:

• Brad Hjelmar talks about how he helped a customer move off of Datadog with stringent inventory analysis –  retaining only one in every five assets. 

• Nima Adib explains how he helped a customer establish a framework for identifying what assets to keep and what to leave behind for a Wavefront migration – leading to a 5x improvement in efficiency.  

• Sarah Hudspeth touches on how she helped a customer trim their spend and improve their telemetry data practices, and onboard Prometheus. In the end, they were able to IPO.
• Ben Raskin breaks down how he helped a customer with a Prometheus upgrade by configuring aggregation rules to help reduce the overall volume of data. 

Check out a transcript of the conversation below! 

​​Sophie Kohler: Are you scared to migrate observability tools? Let’s say your old home was giving you grief with leaky ceilings, noisy neighbors, maybe even a haunted attic. Now, you’re finally considering finding a new home, but the process seems like a lot of time and a lot of work. You have to pack up all of your furniture, belongings, the necessities, the miscellaneous –  like that random box of 400 cords that you’ve had for the past 10 years, but you can’t get rid of it because what if you need it one day?

On top of that, there’s a lot to consider. Like, how much will it cost? And how long will it take?​ Well, just like moving homes, developers might see the process of moving observability tools as daunting. Luckily, Chronosphere offers white glove support every step of the way to help teams navigate what to keep and what to ditch as they move on to their new cloud native observability platform. Some may even call it cathartic.

And best part, you don’t have to migrate in nine months. You can do it in as little as four weeks. Speaking of four, who better to hear from than four of our Solutions Architects on how they’ve helped our customers migrate off of their tool as stress free and painlessly as possible? 

Sophie Kohler: Hey Brad, can you tell us about how you helped a customer move off of Datadog and onto Chronosphere seamlessly?

Brad Hjelmar: Let’s do it! Customers with large Datadog environments often express concern with bringing over the entire hoarder’s house of assets. By doing some stringent inventory analysis beforehand, such as Datadog popularity score, Number of widgets and specific title substrings were able to provide the necessary information to customer teams to make informed decisions on what assets they want targeted for migration.

When customers fully come on board, they’re often able to significantly cut down on their asset inventory, in most cases retaining only one in every five assets. 

Sophie Kohler: Pretty cool. Pretty cool. Pretty cool.

Brad Hjelmar: By doing this analysis, the migration timeline can be significantly shorter than most customers expect, and in the end, teams will actually be happier with their asset inventory in Chronosphere than they were in Datadog. 

Sophie Kohler: Deciding what you need to bring with you can be one of the hardest parts of the process. Hey Nima, can you tell us about another instance when you were able to help one of our customers bring only the assets they cared about with them over to our platform?

Nima Adib: Yeah, I would love to. So, many customers think that they have to drag everything over. So, they look at their existing lists of alerts and dashboards and they think migrating all of that is this monumental task that scares them a bit and they don’t think it’s worth taking on. But that’s not really the case.

We recently helped a customer migrate from Wavefront one of the first things we did was run an analysis on their existing assets That includes dashboards that were just not being visited anymore, or things that were set up as test or duplicate dashboards, as well as alerts that may have been broken in the first place and not firing at all, or going to destinations that don’t exist anymore.

Sophie Kohler: Uh, whoa. So how much were they able to simplify by?

Nima Adib: by following this structure, we were able to reduce the overall number of assets that the customer has to manage by about 50%. They even mention a 5x efficiency improvement in how they retrieve data and view dashboards. 

Sophie Kohler: Hmm, now what about upgrades? Hey Sarah, could you tell us about how you helped a cybersecurity company upgrade from SignalFx and Prometheus?

Sarah Hudspeth: Hi. Yes, I can, Sophie. I like your analogy of the house, sometimes it’s necessary to upgrade to become more energy efficient Getting new windows or installing solar panels, even though it’s a pain in the short term, there’s a lot of long term value in doing that I like to say that’s exactly what we did with this customer.

A few months after we completed the migration work, they were able to ring the bell at the New York Stock Exchange. We helped them migrate all of their assets. We helped them onboard and understand Prometheus, open source mindsets, and data management. We did something, what I call a migration bootcamp, where we really looked at their data and asked if it was providing value.

At the end they got to IPO. They also have great dashboarding experiences. They really understand their data. We were there helping them with custom enablement sessions, and really got to be good partners and lay a firm foundation for work in the future. 

Sophie Kohler: And we have another instance when we were able to help a Chronosphere customer successfully upgrade their tooling, which this time offered OpenTelemetry and Prometheus compatibility. 

Ben Raskin: So this financial services company was managing a very large Prometheus cluster and was experiencing lots of toil and engineering fatigue. When evaluating vendors, one of the key requirements that this customer had was to have a seamless and low effort migration. By providing a completely Prometheus compatible service, Chronosphere was in a position to fulfill this strict requirement. 

Metric ingestion was as simple as swapping out the Prometheus instances with the Chronosphere collector and copying over the same scrape configurations. Once ingestion was complete, the Chronosphere team helped the customer configure aggregation rules using our control plane. 

Once this was complete, we simply uploaded their Prometheus rules, and uploaded their dashboards into the Chronosphere platform. The migration took around six weeks, with most of the time being allocated for teams to verify everything looked correct between the two systems. 

Sophie Kohler: Listen, we get it. Migration seems scary and it’s a reason that sometimes potential customers don’t proceed or start at all. But, it’s easier than you think.And you can’t afford to not switch off your current very expensive observability platform.

We offer a white glove customer service to help teams through the migration process as quickly and seamlessly as possible. If you’re curious to learn more, check out some of the resources I’ve linked in the description below, and let us know what you’d like to hear about in the next episode. See ya!

You can watch the full Chronologues episode here.