Ce diaporama a bien été signalé.
Nous utilisons votre profil LinkedIn et vos données d’activité pour vous proposer des publicités personnalisées et pertinentes. Vous pouvez changer vos préférences de publicités à tout moment.

Datadog: From a single product to a growing platform by Alexis Lê-Quôc, CTO

110 vues

Publié le

By Alexis (https://twitter.com/alq), CTO at Datadog (https://www.datadoghq.com)

Alexis built Datadog's whole infrastructure and team from scratch as a co-founder. From a very small & dedicated team with no experience, he learned step by step to build a complete product ️

He shared with us his experience as a co-founder and CTO building a cloud giant in New York. How do you keep learning, how do you interact with customers & your market to drive your product development, and how do you monitor it all to make you company evolve will be the main topics of his talk.

Publié dans : Business
  • Soyez le premier à commenter

Datadog: From a single product to a growing platform by Alexis Lê-Quôc, CTO

  1. 1. Teaching Datadog new tricks From a single product to a growing platform From early cloud natives to enterprise customers Alexis Lê-Quôc (@alq)
  2. 2. What Datadog is – Modern monitoring & analytics of applications and infrastructure, on-premises and in the cloud – Thousands of customers, from the large enterprise to the startup-in-a-garage – A platform of integrated products
  3. 3. 9 years ago – A handful of people – Limited enterprise software experience – 0 customer – (barely) 1 product: infrastructure monitoring
  4. 4. – Customers – Ecosystem – Products – Ourselves What have we learned in 9 years?
  5. 5. 2010-2013
  6. 6. Who we thought we were... “Datadog is a service for IT, Operations and Development teams who write and run applications at scale, and want to turn the massive amounts of data produced by their apps, tools and services into actionable insight.”
  7. 7. Monitoring: dirty word Cloud: not real Notably absent...
  8. 8. Who we really were... “Datadog is a service (blah blah blah) and that wakes you up in the middle of the night when your cloud infrastructure is on fire.”
  9. 9. SaaS for Monitoring + Cloud
  10. 10. Metrics The Pillars of Observability
  11. 11. 2014
  12. 12. Enterprise users start to show up at AWS Re:Invent, where we have our first booth. Financial industry still cautious about the cloud. Docker hits 1.0 and our Re:Invent session on monitoring “Docker containers” is immediately sold out. Integrations are differentiating, including APM integrations. The sea is changing... Customers Ecosystem Ourselves
  13. 13. – Native docker monitoring. – Deeper cloud integrations and more integrations in general. – Explore APM. – Same inside/inbound go-to-market, even in enterprises What we learned
  14. 14. Metrics The Pillars of Observability
  15. 15. 2015
  16. 16. CapitalOne on stage at AWS Re:Invent: first sign that public clouds mean serious business. k8s reaches 1.0 but the fate of orchestrators are still undetermined. AWS lambda was born, promising but limited. From APM integration to APM prototype Cloud and containers are real Customers Ecosystem Ourselves
  17. 17. – Enterprise still adopt clouds bottom-up thus “land-and-expand”. – Individual users want to learn; strong content push – Containers drive complexity; machine learning to the rescue. – Starting a new product from scratch is hard but a platform is born. What we learned
  18. 18. MetricsTraces The Pillars of Observability
  19. 19. 2016
  20. 20. Deals get bigger and bigger as public cloud continue to expand in the US. Other cloud providers start to play serious catch-up. k8s is winning the roll-your-own orchestrator wars; very young and aspirational. Serverless was born. APM is real, how about logs? More cloud, more containers Customers Ecosystem Ourselves
  21. 21. – Start enterprise sales in the US. Sell inside globally. – Dedicate teams to k8s to stay up-to-date. – Serverless is the new docker; experiment and bet. – APM: it’s all about integrations. – Everybody needs logs. What we learned
  22. 22. MetricsTraces The Pillars of Observability
  23. 23. 2017
  24. 24. Ex-US cloud adoption still trails US by 2 years but global enterprise customers are migrating to cloud. Manager container orchestration is still proprietary (ECS, Fargate); will k8s win there too? Infrastructure + APM + logs: the return of the “single pane of glass”? Global cloud adoption Customers Ecosystem Ourselves
  25. 25. – Replicating the sales model around the world takes time and effort (duh). – APM is GA: java, .net and php integrations have landed. – Log management in the cloud is a tricky business. – Being k8s-native is the best way to support k8s. What we learned
  26. 26. MetricsTracesLogs The Pillars of Observability
  27. 27. 2018
  28. 28. Public clouds have become mainstream, have crossed the chasm. k8s has also won the managed orchestration battle. Serverless is here to stay but may need another generation or two to win. Providing end-to-end visibility will require a true platform, more than a collection of adjacent products. Datadog outside of the datacenter: synthetics Cloud + container + serverless Customers Ecosystem Ourselves
  29. 29. – Try before you buy also works for global enterprise sales. – Starting new products remains hard (organic/acquisitions) What we learned
  30. 30. MetricsTracesLogs The Pillars of Observability Synthetics
  31. 31. – Still a service for customers – Still focus on time-to-value – Still learning What has not changed in 9 years...
  32. 32. Thank you