Live data from Hacker News

Pingfs: A filesystem where data is stored as ICMP Echo packets (2016)

github.com

31–40 of 45 posts

Re: Pingfs: A filesystem where data is stored as ICMP Echo packets (2016)

#31
post #22

Earlier quoted context omitted.

illumos and it’s the programmer’s “Linux only” attitude which irritates me.

I get your sentiment in general but... you're upset because a tool for storing files in IMCP pings isn't available on your OS? Poor you, where else are you going to keep your files!

No, I’m irritated because of the programmer’s attitude, which is quite systemic in programming and has been for a while.

Re: Pingfs: A filesystem where data is stored as ICMP Echo packets (2016)

#32
post #31

Earlier quoted context omitted.

I get your sentiment in general but... you're upset because a tool for storing files in IMCP pings isn't available on your OS? Poor you, where else are you going to keep your files!

No, I’m irritated because of the programmer’s attitude, which is quite systemic in programming and has been for a while.

Well, it's not merited, and you're being rude. I'm saying this as someone who references the POSIX standard several times a week to ensure portability. This kind of tool is a toy project and is by its nature highly dependent on non-portable interfaces.

Re: Pingfs: A filesystem where data is stored as ICMP Echo packets (2016)

#33
post #23

Earlier quoted context omitted.

Your calculation makes sense. However, for those 20 MB in flight, how much data is stored in buffers? I think that could be approximated by comparing ping time with speed of light. e.g. if a 5000 km long link at 10 Gbit/s has latency of 0.0166 s, then you have zero buffering. If you have latency of 0.2 s, then around 2 Gbit of data must be stored in buffers somewhere, making 20 MB in flight only 1% of all data stored…

How about the earth-moon-earth path for 768,000km? EDIT: For the EME path that's around 3 gigs in flight

We've found a viable way to a space elevator: Sell it as a delay-line storage device.

Re: Pingfs: A filesystem where data is stored as ICMP Echo packets (2016)

#35

Earlier quoted context omitted.

nsafs might be another good choice: https://github.com/freedomtools/nsafs

I didn't realize there were so many of these jokefs projects. It's like the software dev equivalent of those silly domain jokes: https://helenkellersimulator.org/ http://motherfuckingwebsite.com/ http://thingsididlastnight.com/

I wrote a joke filesystem that does nothing but provide 1x1 PNG files[0]

(I was bored and asked for dumb filesystem ideas on IRC)

0. https://github.com/sector-f/colorfs

Re: Pingfs: A filesystem where data is stored as ICMP Echo packets (2016)

#36

This is exactly what I needed - not a lot of people are doing eventually inconsistent databases yet. If theres a network partition, shits probably fucked anyways so its best to throw away all your data. Unfortunately the metadata (list of files) is stored in memory, not network. Not quite as unreliable as I would like.

Eventually inconsistent databases could save a lot of the effort developers faced with databases without inconsistency guarantees put into implementing eventual inconsistency at the application level.

Re: Pingfs: A filesystem where data is stored as ICMP Echo packets (2016)

#37

Earlier quoted context omitted.

nsafs might be another good choice: https://github.com/freedomtools/nsafs

I didn't realize there were so many of these jokefs projects. It's like the software dev equivalent of those silly domain jokes: https://helenkellersimulator.org/ http://motherfuckingwebsite.com/ http://thingsididlastnight.com/

Another good one: https://github.com/philipl/pifs

Re: Pingfs: A filesystem where data is stored as ICMP Echo packets (2016)

#38

This reminds me of delay line memory, storing information in the echo of a sound: https://en.wikipedia.org/wiki/Delay_line_memory . LEO had long tubes of mercury for this, not sure I would have wanted to work in that office.

Actually, Engineering Research Associates (predecessor of Univac) built a CPU with mercury delay line memory using acoustic transducers to create a shift register. It did bit-serial binary arithmetic. You could get two logic gates or one latch out of a dual pentode, so a bit-serial ALU could fit easily in a single rack.

I'm not familiar with the overall architecture, though, unfortunately. I do remember having to do a most-significant-bit first, bit-serial, two's complement adder as a homework assignment that was inspire by the machine.

Re: Pingfs: A filesystem where data is stored as ICMP Echo packets (2016)

#40
post #23

Earlier quoted context omitted.

Your calculation makes sense. However, for those 20 MB in flight, how much data is stored in buffers? I think that could be approximated by comparing ping time with speed of light. e.g. if a 5000 km long link at 10 Gbit/s has latency of 0.0166 s, then you have zero buffering. If you have latency of 0.2 s, then around 2 Gbit of data must be stored in buffers somewhere, making 20 MB in flight only 1% of all data stored…

How about the earth-moon-earth path for 768,000km? EDIT: For the EME path that's around 3 gigs in flight

In a small toroidally shaped universe, your radio is also your memory. All emissions eventually return to their origin, but you can reorder memory transmissions by moving while the data is in flight.
Post reply on HN