Lots of nuggets of wisdom in this post, many of them informed directly from experiences in production. I hear plenty of people talking about "building operators" as the next step after Helm charts these days, but haven't seen many dig into the details of what that really means and how to go about making it a reality.
Ah, yeah, you weren't the only one bitten by that. We actually went and changed the Cloudant metering model recently so that you're billed on provisioned throughput rather than total request volume. You get dramatically more predictable billing, with the tradeoff that clients need to handle a 429 Too Many Requests response if the Cloudant instance is under-sized. More here:
IBM actually "cares about the little guy" to an incredible degree these days. The last few years have seen a major shift in buying trends in the enterprise that puts developers in a position of significant influence. We know that if those individual devs have a chance to experiment with technologies as hobbyists and have a positive experience they'll be much more likely to recommend them for the next internal corporate project. Bluemix is the best example of our commitment to this space and the Compose services are a natural complement to that PaaS environment.
A relevant counter-example: the group bringing Compose into IBM is run by the former leadership team at Cloudant (including yours truly). We're incredibly excited to have mrkurt & co. on board; Compose does a great job of filling out the Bluemix services portfolio and gives us a solid platform for launching new offerings going forward.
Yep, Damien's patch greatly improved CouchDB's write throughput for the case where you've got a pool of processes trying to write to the same DB simultaneously.
