Thank you! The hyperscalers are so well positioned that it's hard to not see them leading in M&A. The caveat would be if regulators block Data & AI acquisitions because they're concerned about anti-competitive practices.
Secondly, the data providers that are aggressively expanding into data platforms such as Snowflake, Databricks, and Datadog are likely to pursue a lot of M&A. What are your thoughts on it?
That’s a good question. It depends on what the driver for a M&A is. In my last job we always said. You either do it because you have to much money, or want a strategic benefit from it. Or you are not innovative enough and what do grow that way. The first two are a good sign. The last one is screaming for help. So essentially you need to be the fly on the wall to answer it confidently. :-)
Great work on this—the box of rocks that occupies the space between my ears could only fully comprehend about 25% of it, but I can still appreciate how thorough it was.
Do you think quantum computing will play a role in this space? So-called "quantum supremacy" still seems far off, but it seems like it could revolutionize this space if the stars align
Thanks Jon, certainly a complex space. Quantum is interesting, I think I’ll write a primer on it in the future. At this point, I haven’t studied it enough to give any real insight. Stay tuned.
Amazing article! Wondering if you could expand on the relationship between database systems (e.g. MongoDB, Oracle, etc) and data warehouses/lakes (e.g. snowflake, databricks, etc). My understanding is that both are used for data storage, with the latter more for longer-term storage & analytical purposes (with databases commonly being a source of data ingested). Is this correct?
That's a pretty good way to summarize it. The first thing to call out is that there is some overlap depending on a company's specific architecture. Generally, databases are used for transactional workloads (inputting, updating, pulling data) for a specific application. Data warehouses are used for analytics, and generally store a wider set of data across a company.
Great Primer. Did you ever consider splitting out the storage bucket into primary and backup storage? Two considerations for doing this. First, the two sources provide different types of value to apps looking to extract value (point-in-time vs. historical) and second backup copies (and metadata) are an emerging source of value for both security and AI projects. And while not all unstructured data sources (e.g., data lakes) are “backed up”, “structured” ones such as Snowflake and Salesforce are and both types of data offer copies that can be leveraged.
This is very helpfull to me, in my journey to understand the different sectors I invest in. It's mesmerising how vast the data space is, you managed to create a good overview accompanied by good info graphics.
Thank you for sharing all this information 🙏
I'm looking forward to read more of your articles!
Love this article. The big question is. Who will lead the M&A game? Thoughts?
Thank you! The hyperscalers are so well positioned that it's hard to not see them leading in M&A. The caveat would be if regulators block Data & AI acquisitions because they're concerned about anti-competitive practices.
Secondly, the data providers that are aggressively expanding into data platforms such as Snowflake, Databricks, and Datadog are likely to pursue a lot of M&A. What are your thoughts on it?
That’s a good question. It depends on what the driver for a M&A is. In my last job we always said. You either do it because you have to much money, or want a strategic benefit from it. Or you are not innovative enough and what do grow that way. The first two are a good sign. The last one is screaming for help. So essentially you need to be the fly on the wall to answer it confidently. :-)
Great way to think about it!
The go-to-primer for investors willing to understand how the top players operate across the value chain.
Impressive work, Eric. Thank you for sharing and for the mention!
Thank you my friend!
I thoroughly enjoy your Primers, Eric. Very impressive work. If you ever decide to accept subscribers, i'll be standing in line!
Thanks Magnus!
Nice post with good summary
Really nice one Eric. Thanks so much!
Thanks AJ!
Great work Eric, really enjoyed the clear overview and writing 👌👍
Thank you Stephan!
Great work on this—the box of rocks that occupies the space between my ears could only fully comprehend about 25% of it, but I can still appreciate how thorough it was.
Do you think quantum computing will play a role in this space? So-called "quantum supremacy" still seems far off, but it seems like it could revolutionize this space if the stars align
Thanks Jon, certainly a complex space. Quantum is interesting, I think I’ll write a primer on it in the future. At this point, I haven’t studied it enough to give any real insight. Stay tuned.
Amazing article! Wondering if you could expand on the relationship between database systems (e.g. MongoDB, Oracle, etc) and data warehouses/lakes (e.g. snowflake, databricks, etc). My understanding is that both are used for data storage, with the latter more for longer-term storage & analytical purposes (with databases commonly being a source of data ingested). Is this correct?
That's a pretty good way to summarize it. The first thing to call out is that there is some overlap depending on a company's specific architecture. Generally, databases are used for transactional workloads (inputting, updating, pulling data) for a specific application. Data warehouses are used for analytics, and generally store a wider set of data across a company.
Great primer Eric and thank you for highlighting my work on Datadog
Thank you! And happy to share, it’s great work
Great Primer. Did you ever consider splitting out the storage bucket into primary and backup storage? Two considerations for doing this. First, the two sources provide different types of value to apps looking to extract value (point-in-time vs. historical) and second backup copies (and metadata) are an emerging source of value for both security and AI projects. And while not all unstructured data sources (e.g., data lakes) are “backed up”, “structured” ones such as Snowflake and Salesforce are and both types of data offer copies that can be leveraged.
This is very helpfull to me, in my journey to understand the different sectors I invest in. It's mesmerising how vast the data space is, you managed to create a good overview accompanied by good info graphics.
Thank you for sharing all this information 🙏
I'm looking forward to read more of your articles!
Thanks Robin!
Thank you. That is about as good as it gets. Both detailed and accessible.
Great work here!
Very impressive work, indeed.
Great piece. Thx!