4 ms·

Ops person here, by moving Redis/PostGres to Redis Inc./CrunchyData, does that mean your queries are running over internet? What are security/response time impl

Ops person here, by moving Redis/PostGres to Redis Inc./CrunchyData, does that mean your queries are running over internet? What are security/response time implications of that? I can already see my InfoSec person going "YOU PUT WHAT ON THE INTERNET?"

sgarland 2y ago
I’ve seen entirely AWS workloads that still routed traffic out to the internet and back in. Devs running infrastructure is always an interesting experience.
- nathants 2y ago
  pretty hard to miss this in surprise billing.
  - sgarland 2y ago
    You might be surprised at what companies will let slip through when the AWS bill is in the millions $/month.
    - nathants 2y ago
      true. congress should do something about egress gouging, it’s insane.
gurchik 2y ago
If you’re using TLS what is the concern? I’d be more worried about Internet data transfer costs than that. Latency might be a concern but it’s going to be very dependent on use case.
- stackskipton 2y ago
  It's not encryption but the fact database could be siphoned off by just stealing the credentials and possibly getting one of your IPs whitelisted. If it's inside the network, they have to establish a bridgehead and maintain it which is in theory, more difficult and higher risk of detection.
marksomnian 2y ago
From an infosec perspective, as long as the queries are encrypted (with proper TLS verification), that angle is covered (though there are other considerations about data sovereignity etc.). In terms of response time, that's something you'd need to benchmark for your application - though, given most DBaaSes run in the same major cloud providers are your application, it'll either be the same region or go cross-region via their private backbone, so the latency delta should be manageable. Of course if your app is particularly latency-sensitive for the DB that won't work.
nullspace 2y ago
From latency perspective, not really. As someone who used redislabs at a previous company, the requests got routed through the Private Network proxy (whatever AWS calls it) which minimizes any networking overhead.

sgarland 2y ago

I’ve seen entirely AWS workloads that still routed traffic out to the internet and back in. Devs running infrastructure is always an interesting experience.

nathants 2y ago
pretty hard to miss this in surprise billing.
- sgarland 2y ago
  You might be surprised at what companies will let slip through when the AWS bill is in the millions $/month.
  - nathants 2y ago
    true. congress should do something about egress gouging, it’s insane.

gurchik 2y ago

If you’re using TLS what is the concern? I’d be more worried about Internet data transfer costs than that. Latency might be a concern but it’s going to be very dependent on use case.

stackskipton 2y ago
It's not encryption but the fact database could be siphoned off by just stealing the credentials and possibly getting one of your IPs whitelisted. If it's inside the network, they have to establish a bridgehead and maintain it which is in theory, more difficult and higher risk of detection.

marksomnian 2y ago

From an infosec perspective, as long as the queries are encrypted (with proper TLS verification), that angle is covered (though there are other considerations about data sovereignity etc.). In terms of response time, that's something you'd need to benchmark for your application - though, given most DBaaSes run in the same major cloud providers are your application, it'll either be the same region or go cross-region via their private backbone, so the latency delta should be manageable. Of course if your app is particularly latency-sensitive for the DB that won't work.

nullspace 2y ago

From latency perspective, not really. As someone who used redislabs at a previous company, the requests got routed through the Private Network proxy (whatever AWS calls it) which minimizes any networking overhead.