virtuallynathan

Big fan of this fork, been using it for everything for the last couple of weeks.

Went from codex/claude code -> opencode -> pi -> oh-my-pi

Try Starlink?

Always fun to see one of my professors from the quite tiny, but awesome computer science department at St Andrews on HN!

Sorry, yes, was referring specifically to the Sunil Tagare version.

FYI, this is NOT the cable Facebook is planning to build, this is the dream cable of a submarine cable… enthusiast? From LinkedIn: “To be clear, this is not Meta's plan or map. This is Ver "T" and T stands for Tagare. It's what I think is going to happen to this cable if I was designing it. This is my wish list.”

https://www.linkedin.com/pulse/map-metas-w-cable-sunil-tagar...

This latest revision of Jupiter is apparently 400G, as is the ConnectX-7, A3 Ultra will have 8 of them!

I’ve heard stories from people who have worked at Hanford, and it seems like a lot of that money is being squandered. Excess caution, basically everything is radioactive waste, and just overall wasteful spending for decades.

There’s more than one slang dictionary, it seems… here’s one I found in person at UCSD’s library: https://photos.app.goo.gl/J3rCLtbJMEn4SkKq8

They called out Flexential in the post?

Pretty wild for a single fiber, but super involved, using 5 different doping type amps. Current systems use just EDFAs (Erbium). These guys used Erbium, Thallium, Bismuth, and others.

I updated my map for AWS in Northern Virginia -- came up with 74 buildings (another source says 76, so i'll call it directionally correct). If I scale my sq ft by ~5% to account for missing buildings, we get 11,500,000sq ft in the northern virginia area for AWS.

I'll finish my other maps and share them later...

Well, they tell you for one: https://datacenterpost.com/scaling-up-google-building-four-s... We also have Google Street view, etc.

Of course, it's all estimates. You can get fancier and count generators and transformers and stuff too.

Yep, I am aware, I have a square footage multiplier for their multi-story buildings.

I’m not sure mapping VPCs is super helpful - the physical infra is fairly distinct.

AWS has also disclosed 20 million Nitro adapters have been deployed, so you can do some backwards napkin math from that.

Yep.

Google certainly has more compute density than Amazon, the numbers I was able to find from the local power company was 250MW at Council Bluffs back in 2015 or so.

Amazon builds out 32MW shells, and the most utilized as of 5 or 6 years ago was 24MW or so, with most being much less than that.

To date, facebook has built, or is building, 47,100,000 sq ft of space, totaling nearly $24bn in investment. Based on available/disclosed power numbers and extrapolating per sqft, I get something like 4770MW.

Last I updated my spreadsheet in 2019, Google had $17bn in investments across their datacenters, totaling 13,260,000 sq ft of datacenter space. Additional buildings have been built since then, but not to the scale of an additional 30mil sq ft.

Amazon operates ~80 datacenter buildings in Northern Virginia, each ~200,000 sq ft -- about 16,000,000sq ft total in that region, the other regions are much much smaller, perhaps another 4 mil sq ft. When I'm bored I'll go update all my maps and spreadsheets.

You can just map them all... I have. I also worked at AWS :)

I have mapped every AWS data center globally, and I worked at AWS.

Facebook publishes this data.

Facebook has more datacenter space and power than Amazon, Google, and Microsoft -- possibly more than Amazon and Microsoft combined...

Token buckets have burstiness problems, CAKE lets you run things at ~99%+ of your link speed, assuming a stable medium, without bufferbloat, thanks to the packet pacing it does. CAKE also does several other things using Cobalt (CoDel + BLUE), and clever 8-way set associative hashing that helps out when you have many flows.

I did the math on this back in 2013, based on the annual reported number of hours uploaded per minute, and came up with 375PB of content, adding 185TB/day, with a 70% annual growth rate. This does not take into account storing multiple encodes or the originals.

There's currently a study underway that took 100 people who have been on a low carb high fat diet for, on average, over 4 years. These people had a large increase in LDL when switching to this diet. All participants had CT Angiograms, CAC scans, and other tests at T-0 and after 1 year. Some in the study had LDL levels seen in Homozygous Familial Hypercholesterolemia (>350mg/dL).

A matched control study with this population and the general population is also underway, with the initial results being presented on Dec 8th at the WCIRDCD Conference.

If LDL is independently causal of atheroscelosis/CHD, it should be seen in this population.

I didn’t generally see upstream congestion when I worked on Internet QoE at Comcast, downstream was more common. Cable modems do ACK suppression as well to help there.

It does, you can update it via Satellite. I've done it... (it's a manually triggered operation)

Yep, she was behind regionalization and IPv6 and such. I recall reading the same the the parent comment talks about.

I had some media files I wanted to verify the integrity of -- I was using tdarr.io in a podman container on my local server. I then spun up an m7a.12xl spot instance, mounted my local storage to it via SMBv3. I ran tdarr on that instance, and connected it as a worker node to my home server instance. tdarr ran ffmpeg to decode every video frame to /dev/null, and reported any errors. It was able to process a few thousand FPS or about 2-3Gbps. I found some corrupted files, and was able to restore them from another backup.

There was no practical reason to do this, it cost about $100 to double my processing speed.

I just upgraded to 10GbE service from Ziply Fiber in the Seattle area:

Speedtest.net - ~9.3Gbps symmetric

iperf to Denver: ~9.3Gbps symmetric (~8 threads)

iperf to Minneapolis: ~9.3Gbps symmetric

AWS S3 Download with small files: 1.3Gbps

Usenet download: 7-9.5Gbps

Mounting my ZFS pool to an AWS instance via SMBv3 in us-west-2: 3Gbps (not clear what the limit was here)

GitHub (400MB): 1 thread 400Mbps, 16 threads: 3.3Gbps

Fedora ISO from local mirror: 1 thread: 1.4Gbps, 16 threads: 4Gbps

Will try to test Steam and such.

That one seems a bit pricey, I think the best I’ve seen for CADR per dollar is the Conway 400(S) or Medify MA-112.

The live video encode use-case seems like one that makes sense, and likely requires more bandwidth than TB4 can support.