There’s issues specific to workflows “agents”. For example many requests in an agent are all on top of the same previous results so context (kv cache) needs to be longer, and they use these massive connected nodes with direct nvme to cache the part of the prompt that’s repeatable.
It is about agents in that the design is for long context, many requests where the initial “chunk” is cached but spread across many requests.
They don’t call this out specifically but in the technical details like about the sram, how it’s all interconnected nodes in a pod it’s “designed” for it.
12 hour shifts are 13+ hours of actually being there because of lunch / breaks. For example Genentech in SF / Vacaville are "12 hours paid shifts" with 1 hour mandatory lunch. So no it is not required for an overlap because there i already a required by law break (lunch) that forces overlap (assuming this applies to more states than just CA).
I believe the difference is this is part of GCP and their terms for GA products are that they will support for at least one year on deprecated versions / products and not just instantly remove something.
I tried to do this for my gym. After many back and forth calls with the bank and them telling me they just have to dispute the charge and do something else. Went in person to the bank (WF) and she said there was nothing they could (probably means will) do and I have to get a new credit card).
Thought I would chime in here since I didn't see anyone comment here about gold. We have GCP gold support. They answer all the questions / issues quickly and there aren't any limits to ticket counts I am aware of. During both the outages we have been affected by (~1.5 hour google load balancer outage, >3 hour bigquery outage) in the past few months, the support feels pretty bad even though there is nothing they can do I guess. However during the bigquery one, it was "outage will have an update by some time". There proceeds to not be some update for well past that time. Also 3 hours later its still we will have an update by some time but no more info when both of these were past SLA. Overall feels good and they answer any weird / general questions too which is nice (although sometimes on every reply they say I am going to close the ticket).
I had this same issue (a while ago). I believe the solution was to turn back on icloud photos and then your photos show up in the photos app where you can delete them, and then turn off icloud photos again (convenient)
I have used them for a side project, no issues.
The big reason for me moving from DO to them that I didn't see mentioned is because they have actual private networking.
And also the option for servers with HDD instead of SSD.
It is about agents in that the design is for long context, many requests where the initial “chunk” is cached but spread across many requests.
They don’t call this out specifically but in the technical details like about the sram, how it’s all interconnected nodes in a pod it’s “designed” for it.