It depends on the quality of engineer hiring for the 'team' in question. I reckon they might have pretty solid retention rates for the SEs I'm thinking of ...
20.10.2025 09:51
👍 0
🔁 0
💬 0
📌 0
It depends on the quality of engineer hiring for the 'team' in question. I reckon they might have pretty solid retention rates for the SEs I'm thinking of ...
Yeah, I don't know what to say. It would be nice if it escapes hyperscaler top-down management and gets some real contributions. But at the same time a legacy RDBMS wrapped in another layer is not the MongoDB I want.
If you want on-prem don't forget Percona gives a much better deal for that too. The product name is a bit awkward - Percona Server for MongoDB (PSMDB) - but it's source-on-demand opensource (they can't do better than that due to SSPL), Including the 'enterprise' extra stuff.
A postgres-wrapper is not the MongoDB server I want - the performance doesn't compare. But MongoDB Atlas in the reasonable price tiers doesn't have real performance either imo. Document db DBaaS market competition starts now?
AWS, the owners of "AWS DocumentDB (with MongoDB compatibility)", yesterday joined the "DocumentDB" opensource project, another postgreql-wrapping solution created by MS for the same sort of DBaaS in Azure. And its simultaneously been accepted into the Linux Foundation.
By *someone" you mean you right?
"Vibe ops" is coming, isn't it?
What I mean is: unikernels have a hope to be just-click-purchase-done deployable to any cloud user; databases on modified linux VMs the user has to maintain don't.
With the unikernels the OS is out of the hands of the user. The hypothetical unikernel cloud DBaaS provider takes care of that. A big onus to build right, of course. But none of the end-user cloud customers have to provide a Linux admin.
It's a customized Linux kernel. Real-world adoption would be limited to organizations with really good linux admins in their database team. Other research-proven approaches using better-for-dbs drivers in normal Linux, eg. SPDK, haven't succeeded in the commercial world. Likewise for this, I think.
New blog post on the fun new hardware advancements which databases can leverage for great gains, and why the cloud means it doesn't matter that they exist. 🫠
transactional.blog/b...
Reading Victorian England novels is interesting as they were busily inventing huge chunks of the modern world. Which in many ways was quite stable until the last 20-30 years or so where we've been busily inventing TBD (we can't seem to stabilize on anything)
