Today's Debian technical committee resignation: Ian Jackson

Posted Nov 20, 2014 13:42 UTC (Thu) by deepfire (guest, #26138)
Parent article: Today's Debian technical committee resignation: Ian Jackson

I am grateful for Ian's insistence on hedging against the uncertainties that systemd adoption brings to us.

Let's be honest -- it's not a debate about an init system. Systemd is much more, and aspires for still more.

Systemd seeks a profound change to how Linux systems are composed.

God help us.

to post comments

Today's Debian technical committee resignation: Ian Jackson

Posted Nov 21, 2014 17:26 UTC (Fri) by Kamilion (subscriber, #42576) [Link] (2 responses)

Uh? According to the last graphs of market share I saw, "Linux Systems" are far and above anything else out there due to Android's usage of the Linux kernel.

http://cdn-static.zdnet.com/i/r/story/70/00/008699/meeker...

So, unless the BSD-license proponents at Google go away, I seriously doubt systemd will be able to effect a profound change to how Linux systems are composed.

However, what I actually suspect you meant was GNU/Linux, linux with the GPL3 GNU userspace running glibc and a standard unix-style structure based against the POSIX and FSH specifications.

In reality, these systems are extremely rare when we have supercomputers that have 4096 CPU packages running Single System Image linux, and counted as one computer just the same as the dualcore galaxy S3 in my pocket.

In my *personal* opinion, I am not personally thrilled with 'sticking with POSIX' and feel that there is still 'nothing else' out there that really tries to do their own thing besides maybe Gobo Linux or QubesOS.

I am quite quite thrilled at the benefits that hardware virtualization has brought to the table, as it means I can nest a nonstandard system within a standard system or a standard posix system on top of something else.

Xen on ARM64 blows people's minds when you pull what looks to be an oversized USB stick from your pocket and explain it's fully capable of running nine distributions of GNU/Linux at the same time.

Let's help ourselves and leave God to deal with his own business.

Today's Debian technical committee resignation: Ian Jackson

Posted Nov 21, 2014 23:47 UTC (Fri) by dlang (guest, #313) [Link] (1 responses)

> when we have supercomputers that have 4096 CPU packages running Single System Image linux, and counted as one computer just the same as the dualcore galaxy S3 in my pocket.

supercomputers are not running a single system image, so they would count as hundreds or thousands of computers.

unfortunately for the stats, they only cound people who bother reporting that they are running linux, so if the supercomputer operator doesn't go out of their way to report the thousands of machines, they won't be visible to the stats.

The same thing goes for all the machines that Google runs in their datacenter. There are a lot of them and Google won't tell you how many, but the fact that they are missing from the stats makes a noticeable difference. The same thing applies to msot corporate datacenters and cloud systems, you don't know how many instances of Linux are running, and have no way of knowing.

Today's Debian technical committee resignation: Ian Jackson

Posted Nov 23, 2014 23:40 UTC (Sun) by csamuel (✭ supporter ✭, #2624) [Link]

Just to comment, as I'm someone who's been working in HPC for a while.

You are both right - there are indeed distributed memory systems out there as dlang mentions (the now "traditional" 20 year old Beowulf clusters) and also large Single-System-Image systems that bind lots of NUMA nodes together into a single machine that Kamilion mentions (think SGI's UltraViolet and earlier Altix systems for example).

Of course building a really massive Single-System-Image is very challenging as I believe the Linux kernel is limited to just 64TB of RAM so you can only get larger than that with distributed memory systems. :-(

All the best,
Chris