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

Cloud VM benchmarks 2026

devblog.ecuadors.net

Post reply on HN
Print view View original

91–100 of 165 posts

Re: Cloud VM benchmarks 2026

#91

Post by dabinat »

rr808 wrote: post #4

Still having your own hardware seems so much cheaper. Maybe even just for dev/uat environments? Every big corporate I have worked at has lower cost of capital than Amazon, and yet they want to move to AWS. I just dont understand it.

I think it can make sense for user-facing services. I host my web and database servers with AWS because unmanaged DBs can be a PITA, Amazon is peered with basically everyone, AWS is way more generous with network speeds than many dedicated / colo providers, and it’s easy to scale capacity up and down. Backend servers are hosted with cheaper providers though.

Re: Cloud VM benchmarks 2026

#92

Post by jbverschoor »

sroussey wrote: post #64
api wrote: post #48

Earlier quoted context omitted.

Big cloud is ludicrously expensive. It’s truly amazing. Bandwidth is even worse. It’s like a 10000X markup.

It’s wild that no one knows just how cheap bandwidth really is. AWS pulled one over on people and it’s like the movie studios still demanding 10% of the top for VHS distribution. Today.

That’s with every industry

Make things look like a complicated black box. Make sure it feels scary to roll your own. Hide the core technical skills behind abstracted skills

Re: Cloud VM benchmarks 2026

#94

Post by scorpioxy »

veeti wrote: post #86
nekitamo wrote: post #83

Earlier quoted context omitted.

Which comparable US dedicated server providers do you prefer?

OVH has dedicated in USA and Canada

I've read several reports from customers saying that their customer service is really bad. Difficult to know with online reviews of course. Does anyone have positive stories to share? I am looking at Australian hosts specifically and Hetzner doesn't have any data centers here.

Re: Cloud VM benchmarks 2026

#96

Post by syncordie »

boberoni wrote: post #11

Anyone have experience with Oracle Cloud and ease of moving away? This benchmark seems to recommend Oracle Cloud, but I’ve heard that Oracle has historically used aggressive licenses and legal terms to keep customers locked-in.

We moved about 400 VMs across AWS and GCP to Oracle Cloud spead across 4 regions. Our approach was to first move our applications to rely only on Load Balancers, Block/Object Storage and Compute from the cloud. No RDS, BigQuery etc. This required a slight increase in SREs/DevOps resources.

Once we did this, the move was fairly easy (the exception being having to write our own auto-scaling logic as the built-in one is very limited). Overall we reduced our cloud spend (even accounting for the additional staff) by about 40%. Bandwidth is practically free and you are not limited to specific combos of CPU/RAM (so you can easily provision something with 7 cores and 9 GB RAM). Another big factor for us was that compute costs in OCI don't vary by region.

I will not recommend using any other managed services from OCI besides the basics (we tried some and they are not very reliable). We've seen minor issues in Networking periodically (Private DNS, LBs or interconnectivity between compute instances), but overall I would say the switch has been worth it.

Re: Cloud VM benchmarks 2026

#97

Post by close04 »

jiggawatts wrote: post #3
nixgeek wrote: post #2

Genoa was a big leap from Milan. Turin is a huge leap again. AMD really is doing spectacularly well at the moment. Kudos to Lisa Su and the team.

> Kudos to Lisa Su and the team. They're a typical hardware maker unable to focus on software, which is why NVIDIA is now a multi-trillion dollar corporation and AMD is "just" a few hundred billion. They've focused too much on CPUs and completely dropped the ball on AI and compute accelerators. It's especially sad considering that the MI300 and related accelerators on paper are competitive with NVIDIA hardware, it's…

AMD has to fight both Intel and Nvidia in the market. It chose to take on Intel and clearly it was a wise decision. You can’t win if you fight every battle at once against much stronger opponents.

And don’t get me started on the valuation of companies riding the AI bubble.

Re: Cloud VM benchmarks 2026

#98

Post by justinclift »

darkwater wrote: post #87
zackify wrote: post #20

I just ran some massive tests on our own CI. I use AMD Turin for this on gcp, which was noted as one of the fastest ones in the article. The most insane part here is that the AMD EPYC 4565p can beat the turin's used on the cloud providers, by as much as 2x in the single core. Our tests took 2 minutes on GCP, 1 minute flat on the 4565p with its boost to 5.1ghz holding steady vs only 4.1ghz on the gcp ones. GCP charges…

Yeah I expected this benchmark to include hosted "metal" hardware with the "per instruction cost" benchmark to see how provider like Hetnzer fare against classic AWS VMs. It's a bit apple to oranges I know, but I think nowadays is what most people compared pure performance cost are interested in. I'm not going to migrate from AWS VMs to GCP or Hetzner VMs, but I might be open to Hetzner hosted servers instead for a m…

> ... but I might be open to Hetzner hosted servers instead for a massive enough cost reduction.

Don't use Hetzner for anything actually important to you. :(

As to why: https://news.ycombinator.com/item?id=45481328

Re: Cloud VM benchmarks 2026

#99

Post by dijit »

PaulKeeble wrote: post #32

A few comparisons just from a gaming PC with a 9800X3D (8C16T 5.2 Ghz Boost). 7 Zip benchmark 9800X3D 130 GIPs compression, 134 GIPs decompress. C8A 21577 MIPs (21.5GIPs) compression, 9868 MIPS decompression (9.9GIPs). Geekbench 5 9800X3D 16975 multithread, 2474 single thread C8A 4049 multithread, 2240 single thread A desktop class CPU is definitely quicker single threaded and multithreaded, no surprises there most o…

Thanks for the benchmarks!

I’d only add that its very common for gaming computers to be screaming fast- moreso than workstation machines or servers which are a bit more conservative with performance and emphasise correctness (slower cores, slower ram, more ECC). Its not a lot, but it can feel annoying when you sit on a company issued workstation that cost €10,000 but get worse performance than a €2,500 gaming computer.

Re: Cloud VM benchmarks 2026

#100

Post by jurgenburgen »

BeastMachine wrote: post #78
deaux wrote: post #74

Earlier quoted context omitted.

Not from people using that same phrasing twice within a few sentences. """ No warning. No traffic spike. Just… more money gone. That’s when I finally looked at Hetzner. I’ve seen too many backend systems fail for the same reasons — and too many teams learn the hard way. So I turned those incidents into a practical field manual: real failures, root causes, fixes, and prevention systems. No theory. No fluff. Just produ…

Repetition is a very common tool in writing (ie 'I have a dream'). I'm just irked that it's being called out for AI slop because "I feel it in my bones!!" There's a good chance it was written using AI -- should that matter? If the content is wrong/sucks, say that instead. If you're going to dismiss all AI assisted writing: good luck in the next decade.

Reading the same (very annoying marketing blog) style of writing gets old fast.

It’s like suddenly all memes are just the same meme and nobody makes their own memes because “AI does it better”.

The style of writing is an intrinsic part of communication, if you can’t critique that then what is content? We’re not machines sharing pieces of data with each other.

Post reply on HN