Can Azure/AWS/GCP kill likes of DataBricks/Confluent/Snowflake over time?

Dec 19, 2020 12 Comments

Trying to understand the long-term vision of these independent companies who builds on data sets originally living in some other cloud company.

Are they under constant threat that an in-house solution by these big powerhouses can kill their business? As a client/engineering head, why would i want to spend money on two different platforms if one cloud provider is giving a similar solution? Think cost, security etc.

What am i missing that's so unique to these companies that big players cant build what they have built? #tech #bigdata #paas #iaas

116 PARTICIPANTS SELECT ONLY ONE ANSWER
VOTE VIEW RESULT

comments

Want to comment? LOG IN or SIGN UP
TOP 12 Comments
  • I mean, they could’ve ‘killed’ Snowflake but that hasn’t happened.

    Customers are getting smarter and want multi-cloud strategy for their data. Databricks offers that. On top of that, their Spark is a 10-100x better/faster than what anyone else ever promises to offer you
    Dec 19, 2020 4
    • i agree with your assessment but that problem should be for the core foundational components like the persistent storage (SQL servers, cosmos db).

      The more stateless solutions should not be relatively a big deal to move when the business decides to move them. For more stateless components i would be more afraid of the cost in dollars/quality/security etc.
      Dec 19, 2020
    • IBM
      fangchod

      Go to company page IBM

      fangchod
      Even IBM offers what Snowflake offers but Snowflake executes it a bit better and is nice and shiny.

      It's only a matter of time before one of the big companies copies what they do and gives their customers an offer they can't refuse.

      Back in 2018 we said Snowflake is overvalued, because we know this market very well. Fast forward today, imo it's not even worth 1/10th of its valuation. The stock market is driven by retail investors feeling FOMO after seeing their friends betting on TSLA, AMD, BTC and quadrupling their money in a few months.
      Dec 19, 2020
  • Amazon
    yolo_msft

    Go to company page Amazon

    yolo_msft
    Confluent is a joke (currently working there)
    Dec 19, 2020 1
  • Amazon
    nunyabeeswa

    Go to company page Amazon

    nunyabeeswa
    Customers will pay for domain knowledge, as all cloud providers become more reliable. For example, databricks for spark, mongo atlas etc. Elastic could have done well here but screwed it up by pouting like a child. The other thing they will pay for is inter operability. A medium sized company cto s wet dream is to move from one cloud to another within a few month s and show that they saved x MM a year for the next 2 years.
    Dec 19, 2020 0
  • I can't speak for the other two, but for confluent I think it comes down to if the other companies have a will to invest the resources to compete technologically. I don't think we've built a good enough technological differentiator at this point to prevent one of a/g/m from catching up rather quickly. But I know we're trying with things like changing our open source license (now confluent community license)

    One other consideration for a/g/m is if it's worth it to take the chance of building their own when they already get large amounts of revenue from partnering with each of those companies.

    From my experience, things like MSK aren't nearly as good as what we offer, but it's a decent enough facsimile that it eats at our sales. I don't see it as a reason for churn, just making things harder to land. It'll take a much larger investment from aws to really hurt confluent.

    In general I'm bullish on confluent, this is something to be concerned about, but it's not as big of an issue as two years ago imo.
    Dec 19, 2020 0
  • This comment was deleted by original commenter.

    • these companies i mentioned are not a backup store right (they are more stateless/ephemeral entities)? how do these companies help in disaster recovery?
      Dec 19, 2020
    • At what scale is that true? I haven't seen that at 50M revenue, 100M, 20B... All just used AWS or were in the process of moving from something else to AWS.
      Dec 19, 2020