Hacker Newsnew | past | comments | ask | show | jobs | submit | maybeben's commentslogin

feel lucky. s3 suspend quit working on my thinkpad in -CURRENT some months ago after having worked for like a decade. i didn't notice until i pulled a molten hot slab of locked up laptop out of my bag


The only laptop I ever had in my entire life where sleep works is my current XPS 17 running win10.

I want to update my hardware to a Lenovo. Not looking forward to new "sleep won't work no matter what you try" adventures.

That sht is like printers: should always work, never does.


Weird. I haven't had a problem with my System76 laptops.


I think it's highly dependent on the particular hardware.

I have two HP EliteBook G8s, one AMD one Intel. Both work perfectly with Linux, although they don't support S3, only the "modern" standby.

The AMD one randomly craps out if suspended in Windows. No idea what it does, but it gets extremely hot and doesn't respond to anything but a hard reset. Sometimes it reboots on its own. Maybe because it gets too hot? No idea.

The Intel one also sometimes gets fairly warm for no reason (I keep it up to date, so it can't be random updates - also happens in my bag, so while unplugged). But sleep now mostly works fine. For the first year or so it would sometimes wake up with a garbled screen.

Those are both your regular basic enterprise laptops, no dedicated GPUs or anything fancy.


The thing is that modern hardware is super complicated and you have to have a good vendor that specifically targets the OS you want.


Huh, interesting. Are you publishing the results of that testing somewhere?


Mac OS was awful. OS X was amazing. macOS feels like increasingly typical design-by-committee rudderless crapware by a company who wishes it didn't have to exist alongside iOS.


RPKI is obviously half-baked scareware, but why anyone cares what technology the US government, much less "white house" thinks should be used is beyond me


i mean, they also have one of the largest ceph deployments. anything is scalable with no budget.


slide 22 states that the cost is 1 CHF/TB/month (on 10+2 erasure coded disks), though it would be interesting to do a breakdown of costs (development, hardware, maintenance, datacenter, servicing, management, etc..)


1 CHF/TB/month is a bit expensive for storage at that scale, so it would definitely be interesting to see what they're spending the money on and what they are (and aren't) counting in that price.


Tape backup, accessibility, networking, availability... At 1CHF/TB that's a lot better than my local university still charging >100x that for such services internally


Economies of scale in storage are significant. Also, I don't know why you put up with your university charging 100x that when you can store things on AWS for $5-10/TB/month (or less). That comes with all the guarantees (or more) of durability and availability you get from the university.


Don't forget bandwidth costs at this scale are significant to say the least.


I assume most of that exabyte is stored in 2-3 datacenters, in which case the bandwidth cost is actually relatively small. Downloading it would cost a fortune (or take an eternity), but if it stays in the datacenter (or stays in AWS), bandwidth is cheap.


No budget often tags along with no accountability


They probably consume Panasas, IBM, DDN, and BeeGFS gear and licensing too.


Nop.

Most internal data is spread between Ceph and home-made distributed storage system named EOS (https://indico.cern.ch/event/138478/contributions/149912/att...) running over commodity hardware.

The only commerical-backed storage system is the long term storage tape system. Still it has an home-made overlay API over it to interface with the rest of the systems.


Good god no. Nowhere near anything so crass. CEPH and EOS all the way


All their important "administrative" stuff (think Active Directory/LDAP user database, mail and other essential services) run on proprietary storage systems from a commercial vendor (not IBM though), with enterprise support and all.

At least that was the case a few years ago when we last talked to one of the heads of IT at CERN, but I guess it hasn't drastically changed


and still not funny


looking forward to a lengthy blog post blaming BGP and/or some other operator


you can do two solid/high impact boards on indigo 2. each will be a separate X display


maybe everyone in the world using the same handful of resolver addresses isn't so great for fault tolerance


DJB was right. We should have used Curve 25519 for encryption in DNS itself. Then we wouldn't have needed DoT nor DoH, and we'd still have had a pretty good measure of privacy relative to ISPs and eavesdroppers.


I use this facility on FreeBSD (dcons) to do debugging. Works great. There's a knob to explicitly turn on DMA, though. It looks like Linux has something similar. Is that not sufficient?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: