5 ms·

Don't. It doesn't make sense to try to jump into an area rapidly being commoditized. At most, rent bare metal/dedicated servers rather than VPS. > Has anyone d

Don't. It doesn't make sense to try to jump into an area rapidly being commoditized. At most, rent bare metal/dedicated servers rather than VPS. > Has anyone done this and what was your experience? It works as expected. Big cost/time upfront to get all the equipment setup, then things mostly just worked. The primary challenge is transferability and standardization. > How do you select a colo and what do you look for? Bandwidth availability. Proximity to the team supporting it. Reputation among customers. > How do you manage the hardware and how much savings in time/$ is there really? Whatever custom stack is built in the colo must be well documented, and have solid maintenance schedules, and change control procedures. The cost savings exists, but it's never as much as you think. It's difficult to articulate the specific drawbacks. It's more organizational than operational. Imagine that you have a server dropping at random times. The logs say something ambiguous, and now you have to become full time sysadmin to some elusive problem. No one in your organization will want to deal with these sorts of issues, nor will they want you to burn time on it. There will inevitably be other tasks higher on the list. Operational issues gradually build up like other forms of technical debt. No one receives applause for updating the server BIOS over the weekend. Operational discipline/maintenance inevitably becomes last priority.

User23 5y ago
The major advantage of owning your own servers is that you, well, own them. Ideally you'd rent space at a carrier hotel like One Wilshire where you have access to numerous providers in the meet me room. That way if whatever it is you're doing suddenly becomes unfashionable, you can't have some activist cloud provider decide to destroy your business on a whim.
zerkten 5y ago
What you describe only becomes a bigger problem with staff turnover. It gets to a point where fewer people know the ins and outs of the setup and are prepared to maintain it. You end up burning out a few people, or bringing in people just to deal with low-level operations issues that were created by the move away from the VPS-level of operations. The question to ask, is it strategic to move to running our own servers? If I had a large, stable, and boring set of apps which consistently didn't take advantage of "cloud features", then it might make sense to consider moving to colocation because the only option to improve profitability may be to reduce cost. That assumes that major changes won't be made to the app that require cloud features. This is rarely the case. The opportunities seem to only appear at huge scale.
toast0 5y ago
> It's difficult to articulate the specific drawbacks. It's more organizational than operational. Imagine that you have a server dropping at random times. The logs say something ambiguous, and now you have to become full time sysadmin to some elusive problem. No one in your organization will want to deal with these sorts of issues, nor will they want you to burn time on it. There will inevitably be other tasks higher on the list. Operational issues gradually build up like other forms of technical debt. No one receives applause for updating the server BIOS over the weekend. Operational discipline/maintenance inevitably becomes last priority. This sort of thing happens in dedicated hosting too. Unless you want them to just swap a box (which is OK I guess), chances are their support team doesn't know what's going on, and isn't going to spend all day diagnosing your weird issue that only happens sometimes. So someone on your team needs to do it. I don't have enough experience in cloud hosting to know, but I'd guess it happens there too; although I'd guess if you see weird issues, swapping to a new VM is standard operations there.

User23 5y ago

The major advantage of owning your own servers is that you, well, own them. Ideally you'd rent space at a carrier hotel like One Wilshire where you have access to numerous providers in the meet me room. That way if whatever it is you're doing suddenly becomes unfashionable, you can't have some activist cloud provider decide to destroy your business on a whim.

zerkten 5y ago

What you describe only becomes a bigger problem with staff turnover. It gets to a point where fewer people know the ins and outs of the setup and are prepared to maintain it. You end up burning out a few people, or bringing in people just to deal with low-level operations issues that were created by the move away from the VPS-level of operations. The question to ask, is it strategic to move to running our own servers? If I had a large, stable, and boring set of apps which consistently didn't take advantage of "cloud features", then it might make sense to consider moving to colocation because the only option to improve profitability may be to reduce cost. That assumes that major changes won't be made to the app that require cloud features. This is rarely the case. The opportunities seem to only appear at huge scale.

toast0 5y ago

> It's difficult to articulate the specific drawbacks. It's more organizational than operational. Imagine that you have a server dropping at random times. The logs say something ambiguous, and now you have to become full time sysadmin to some elusive problem. No one in your organization will want to deal with these sorts of issues, nor will they want you to burn time on it. There will inevitably be other tasks higher on the list. Operational issues gradually build up like other forms of technical debt. No one receives applause for updating the server BIOS over the weekend. Operational discipline/maintenance inevitably becomes last priority. This sort of thing happens in dedicated hosting too. Unless you want them to just swap a box (which is OK I guess), chances are their support team doesn't know what's going on, and isn't going to spend all day diagnosing your weird issue that only happens sometimes. So someone on your team needs to do it. I don't have enough experience in cloud hosting to know, but I'd guess it happens there too; although I'd guess if you see weird issues, swapping to a new VM is standard operations there.