Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Is Northern Virginia still the least reliable AWS region?

statusgator.com

Post reply on HN
Print view View original

81–84 of 84 posts

Re: Is Northern Virginia still the least reliable AWS region?

#81

Post by paradox460 »

transcriptase wrote: post #55
paradox460 wrote: post #6

Earlier quoted context omitted.

We vibe code everything in flavor of the month node frameworks, tyvm, because elixir is too hard to hire for (or some equally inane excuse)

I look forward to the eventual launch of a new and improved version of your app using electron. What’s the point in having 64 Gb of DDR5 and 16 cores @ 4.2 GHz if not to be able to have a couple electron apps sitting at idle yet somehow still using the equivalent computational resources of the most powerful supercomputer on earth in the mid 1990s.

We also plan to incorporate a full local llm, to ensure we fill the memory up. It will be used to direct people to our online knowledge base, which will always be empty

Re: Is Northern Virginia still the least reliable AWS region?

#82

Post by paradox460 »

DANmode wrote: post #10
paradox460 wrote: post #6

Earlier quoted context omitted.

We vibe code everything in flavor of the month node frameworks, tyvm, because elixir is too hard to hire for (or some equally inane excuse)

I agree with your post conceptually. However: Don’t underestimate community support (in the areas you’re likely to want it) when comparing development stacks.

Conversely, a community means nothing if they flit from one "best practice" to another

Re: Is Northern Virginia still the least reliable AWS region?

#83

Post by transcriptase »

paradox460 wrote: post #81
transcriptase wrote: post #55

Earlier quoted context omitted.

I look forward to the eventual launch of a new and improved version of your app using electron. What’s the point in having 64 Gb of DDR5 and 16 cores @ 4.2 GHz if not to be able to have a couple electron apps sitting at idle yet somehow still using the equivalent computational resources of the most powerful supercomputer on earth in the mid 1990s.

We also plan to incorporate a full local llm, to ensure we fill the memory up. It will be used to direct people to our online knowledge base, which will always be empty

Make sure another LLM summarizes pages upon loading, but doesn’t load any content before that completes. Each page should have a few megs of JS tracking scripts siphoning the users CPU to create massive logs on AWS that nobody will ever use to improve anything.

Oh and put everything behind the strictest cloudflare settings you can, so that even a whiff of anything that’s not a Windows 11 laptop or iPhone on a major U.S. network residential or mobile IP gets non-stop bot checks!

Re: Is Northern Virginia still the least reliable AWS region?

#84

Post by Manouchehri »

htrp wrote: post #73
Manouchehri wrote: post #53

Earlier quoted context omitted.

Yeah, I was often the single source of reporting Claude outages (or even missing support completely) on less commonly used Amazon Bedrock regions.

Which regions were you using ? ( Thought claude had global inference support that routed to all regions)

I believe I was using us-east-2.

In the early days of cross-region inference, less people were using it, and there was basically no monitoring (and/or alerting) on Amazon's side.

The cross-region and global inference routing is... odd at times.

Post reply on HN