3 ms·

I'd love to hear feedback about any aspect of Thiicket, including the website and pricing, good or bad

masukomi 6y ago
I'm with Junon. I don't get why I'd want _REDIS_ to be remote, then i loose all the speed benefits of redis. Also i don't get why i need my redis cluster to scale. The only case where that seems like a need is if I'm caching global variables which is widely frowned upon. Everything else, there are caching + eviction strategies to handle this. If i've got massive traffic such that my redis cache is constantly churning (and thus not providing value) then couldn't i just spin up more redises? redi? along with all the new machines i'm surely needing to spin up. If my engineers are being paged to spin up more redis instances (or aNYTHING) then they've failed in their autoscaling configurations, and it's good they got called, and hopefully they'll rethink their architecture and scripts so that they never get called again. I'm not a devops guy so maybe i'm missing something obvious. On the other hand, I'm not a devops guy and I'm not _seeing_ an "obvious" win here. Either it actually doesn't meet my needs or I'm ignorant about something and the marketing isn't meeting my needs. Side note: not a devops guy but my team has hundreds of AWS machines and lots of redis usage.
- stephenr 6y ago
  The only thing I can imagine they might be offering is redis that scales without the drawbacks of the official redis cluster sharding. I don't know if that is what they're doing, but it's one possible thing they could be doing "better" (as in its an area that could be improved, not necessarily that it's fundamentally or technically possible/easy)
- pier25 6y ago
  > I don't get why I'd want _REDIS_ to be remote, then i loose all the speed benefits of redis If your cloud function is running in say East Asia and your Redis instance is nearby, it will be much faster than querying your DB in Virginia.

I'm with Junon. I don't get why I'd want _REDIS_ to be remote, then i loose all the speed benefits of redis. Also i don't get why i need my redis cluster to scale. The only case where that seems like a need is if I'm caching global variables which is widely frowned upon. Everything else, there are caching + eviction strategies to handle this. If i've got massive traffic such that my redis cache is constantly churning (and thus not providing value) then couldn't i just spin up more redises? redi? along with all the new machines i'm surely needing to spin up. If my engineers are being paged to spin up more redis instances (or aNYTHING) then they've failed in their autoscaling configurations, and it's good they got called, and hopefully they'll rethink their architecture and scripts so that they never get called again. I'm not a devops guy so maybe i'm missing something obvious. On the other hand, I'm not a devops guy and I'm not _seeing_ an "obvious" win here. Either it actually doesn't meet my needs or I'm ignorant about something and the marketing isn't meeting my needs. Side note: not a devops guy but my team has hundreds of AWS machines and lots of redis usage.

stephenr 6y ago
The only thing I can imagine they might be offering is redis that scales without the drawbacks of the official redis cluster sharding. I don't know if that is what they're doing, but it's one possible thing they could be doing "better" (as in its an area that could be improved, not necessarily that it's fundamentally or technically possible/easy)
pier25 6y ago
> I don't get why I'd want _REDIS_ to be remote, then i loose all the speed benefits of redis If your cloud function is running in say East Asia and your Redis instance is nearby, it will be much faster than querying your DB in Virginia.