Quantcast
[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr / vt ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

/vt/ is now archived.Become a Patron!

/g/ - Technology


View post   

[ Toggle deleted replies ]
>> No.75222698

>>75222646
>tarball
in
>2020
yikes

>> No.75223345

>>75222646
I have sowm on my Ubuntu. It works and I like it. I will most likely fork it and if I have time I'll rewrite it in Rust and extend it. Also, I'll mostry likely try to implement mutlithreading for the wm to be faster (the same way as osu!lazer does multithreading).

>> No.75224971

>>75222646
post feet

>> No.75225013

>>75222646
>uptime: 31 min

every time

>> No.75225137

>>75225013
The significance being what, exactly? Everyone always seem to be on about uptimes. Why is that?

>> No.75225331
File: 28 KB, 807x475, VMS_Uptime.png [View same] [iqdb] [saucenao] [google] [report]
75225331

>>75225137
Few understand what real reliability is about today.

>> No.75225942

>>75225831
Because life is too short for endless reboots.
Have you considered you have only about 20000 days left in your life?

>> No.75225960

>>75225942
>Because life is too short for endless reboots.
>if you're not leaving your machine on for years you're endlessly rebooting it
Do all your real life conversations go this way as well?

>> No.75225985

>>75225981
What was your point again? Something about home peecees needing to left on for decades at a time? I recall asking you why that was the case.

>> No.75226082
File: 78 KB, 768x1024, 1579153102174m.jpg [View same] [iqdb] [saucenao] [google] [report]
75226082

>>75225245
It'll be documented on the website for whatever you're downloading.

See:
https://xgboost.readthedocs.io/en/latest/build.html

>> No.75227757

>>75226082
Damn, you can hard wire yourself to the internet now, shits whack yo!

>> No.75227762

>>75225331
15 years of uptime is just a bragging number, in real world scenarios that usually adds way more problems than it's worth, and huge majority of systems like that are either obsolete, not used anymore stuff that is left running by sysadmins for fun or stuff that was left running for so long that everyone is just afraid to touch it, for example because people who set it up and actually know how this shit works no longer work there/are retired/are dead, which is the case with some ancient systems, for example some ancient financial mainframes

you cannot prepare for everything - and are you sure that something that ran for 15 years non stop will boot again? can you troubleshoot it if it wont? system like that is usually years out of support, unpatched, spare parts are hard to come by, et cetera.

planned maintenance breaks are very important - reliability isn't your system running for 15 years straight, it's running 100% of the time when you want it to run.

>> No.75228115

>>75227762
Take a closer look. It is for VMScluster, aka VAXcluster. These machines were not in your basement, rather in a computer centre with people who knew what they were doing. These machines were famous for reliability and support was up to the same standard.

Somehow people have accepted that Win servers go down continuously. The severs we have at work are rebooted every night. People bought VAXen to avoid that.

>> No.75229849

>>75229775
oh and one more real life example but this time not from my - your stuff is fully redundant, there's union strike. someone just pulls off power cords for few critical racks.
hell, maybe someone unplugs wrong machine by mistake.

will it boot?

>> No.75229926

>>75229775
>i seen temps rising inside datacenter to 50 celsius in 5 minutes after one of our chillers went down due to blown compressor, cutting our ac output in half
This is why a university data centre manager was proud to show me that his AC could make snow. Halving his cooling wasn't a biggie for him.

>> No.75230387

>>75230136
20 year old machines are common in defence and aerospace, some telecom and certain embedded stuff. You start a project and make sure the suppliers guarantee decades of supplies. VME bus is still popular, just look up how old that one is.
Uptime requirements in defence and aerospace is demanding and any failure will set off an avalanche of alarms. You really don't want aircrafts in holding patterns until they drop out of the sky while you debug a fancy x86 server maintained and defined by clowns who never bothered to check why people pay obscene amounts of money for big iron and mainframes and just as much in annual licenses.

>> No.75230497

>>75222646
this looks like shit faggot
most likely works like shit, too
stop posting shit and shitting up my board

>> No.75230522

>>75222646
TOTAL WASTE OF TIME

>>
Name (leave empty)
Comment (leave empty)
Name
E-mail
Subject
Comment
Password [?]Password used for file deletion.
Captcha
Action