nerdculture.de is one of the many independent Mastodon servers you can use to participate in the fediverse.
Be excellent to each other, live humanism, no nazis, no hate speech. Not only for nerds, but the domain is somewhat cool. ;) No bots in general. Languages: DE, EN, FR, NL, ES, IT

Administered by:

Server stats:

1.2K
active users

#systems

5 posts5 participants1 post today

💾 Rebooting Petabyte Control Node 💾

am rebooting one of the control nodes for a petabyte+ storage array, after 504 days of system uptime..

watching kernel log_level 6 debug info scroll by on the SoL terminal via iDrac..

logs scrolling, the array of SAS3 DE3-24C double-redundant SFF linked Oracle/Sun drive enclosures spin-up and begin talking to multipathd...

waiting for Zpool cache file import..

waiting.. 131 / 132 drives online across all enclosures.. hmm.. what's this now...

> transport_port_remove: removed: sas_addr(0x500c04f2cfe10620)

well ffs 😒

> 12:0:10:0: SATA: handle(0x0017), sas_addr(0x500c04f2cfe10620), phy(32),

oh, that's a SATA drive on the system's local enclosure bay for scratch data, it's not part of the ZFS pool.. 😌

next step, not today, move control nodes to a higher performance + lower wattage pair of FreeBSD servers 💗

💝 OSS Armv9.2 Motherboard 💝

Radxa Orion O6 ... I must have one. I WILL have one!

The first ARMv9.2 open-source motherboard, designed for ai computing and engineering.

Cix P1 SoC CPU
- 4x Cortex-A720 (big)
- 4x Cortex-A720 (med)
- 4x Cortex-A520 (little)
- 12MB Shared L3

Mem, I/O, Net
- 64GB LPDDR5 RAM
- 4x display outputs
- 2x 5GbE networking
- PCIe Gen4 x8 lane (x16 physical)

GPU, NPU
- Arm Immortals: G720 MC10
- Hardware‑based Ray‑Tracing
- OpenGL ES3.2, OpenCL 3.0, Vulkan 1.3
- 30 TOPs, INT4, INT8, INT16, FP16, BF16, TF32

- docs.radxa.com/en/orion/o6/get
- arace.tech/products/radxa-orio

#Zoomposium with #Dimitri #Coelho #Mollo (Assistant Professor in #Philosophy of #Artificial #Intelligence):
“How intelligent is #artificial #intelligence?”

His #research focuses on # epistemological #questions within artificial intelligence and #cognitive science and seeks ways to improve our understanding of #mind, #cognition and intelligence in #biological and artificial #systems.

More at: philosophies.de/index.php/2023

Or: youtu.be/9xrsxXrXjQw

Excellent Sunday afternoon read from Professor @drmichaellevin for Noema Magazine that takes a look at the metaphors we use to distinguish between organic and non-organic beings and challenges some of the assumptions around what we consider to be machines and/or living things.

Very much in the style of Donna Haraway, he advocates at once for #pragmatism, for empirically testing the methods we use for interrogating systems that imbricate the organic and the machine and to keep an open mind when categorising which is which.

For fans of Douglas Hofstadter, #cybernetics, #systems and #ConsequentialCategories.

noemamag.com/living-things-are

NOEMALiving Things Are Not Machines (Also, They Totally Are) | NOEMAOur formal models of life, computers and materials fail to tell the entire story of their capabilities and limitations.

#Zoomposium with Dimitri Coelho Mollo (Assistant Professor in Philosophy of #Artificial #Intelligence):
“How intelligent is #artificial #intelligence?”

His #research focuses on # epistemological #questions within artificial intelligence and #cognitive science and seeks ways to improve our understanding of #mind, #cognition and intelligence in #biological and artificial #systems.

More at: philosophies.de/index.php/2023

Or: youtu.be/9xrsxXrXjQw

Replied to amen zwa, esq.

@AmenZwa I'm puzzled that you don't mention Erlang. I'm not a programmer, but I follow the discussions. Maybe I'm completely out of date. I thought WhatsApp was built by 5 programmers in Erlang. Then completely rewritten when it was purchased by a large corp. Surely the importance of C in stability is due to the corporate infrastructure. I would have thought the brightest people would use the best new tools for prototyping and proof-of-concept for a new business, then sell it and move on. I'd hope that at least one Ukrainian group is using the Erlang infrastructure to program swarms of drones. I thought #Erlang was good for small and large #systems programming.

There are two kinds of #systems #programming: small and large. By "small", I mean real-time control system, embedded software, or a device driver, the kind that people smarter and ballsier than I would happily write in hexadecimal. By "large", I mean operating system, communications protocol, compiler, database engine, and so on.

It is a fool's errand to design a new language that might replace C in the "small" space. For this type of software, assembly is too craggy, and C++ is too lumpy. Only Forth comes close to C, here.

And it is a herculean task to design a new language that might displace C++ in the "large" space. Countless new languages have tried, and only a few saw some success: Go and Rust, and perhaps Swift, Zig, Nim, and Odin.

This duopoly of the C family is both good and bad. This dominance is good, because it engenders stability. But stability in IT is an illusion. On the other hand, this dominance is bad, because it locks in the adopters and locks out the contenders. And there have been countless examples in IT of this lock-in/lock-out, duo-face problem, over the past 70 years.

What are the alternatives? Well, many arguments could be made in favour of every compiled language in existence. But the two simple choices are these:

• For the "small" space, there are plenty of cases where Forth is a good choice.
• For the "large" space, where safety and security are paramount, the ML family of languages—Standard ML, OCaml (including F# and Reason), and Rust—offer significant advantages over the C family.

Every modern compiled language is an admixture of the C and the ML families—like Rust. Yet, "new" languages keep cropping up, daily.

👩🏻‍💻 Big Brain Computer Parts 👩🏻‍💻

- "Build Day Monday Funday, Yet Another Machine Intelligence System [YAMIS]"
- Season 2, Episode 4: Mostly Modern Systems and Hardware Engineering

Despite some goon sending this kinda pricey Ice Lake Xeon through international post, wrapped up in two thin layers of non/anti-static bubble wrap shoved into a Tyvek shipping bag (not a box!), it arrived yesterday and seems to be in decent shape after a trip from Australia to the middle of America.

So, what one does with a NOS (new old stock) bit of hardware which needs to be perfect in order to function, we inspect with macro photos and various image filtering methods to identify any potential flaws.

CPU: Intel Xeon 8370C
Spec: 32 cores, 64 threads, 2.8GHz base clock
Dimensions: my favorite general purpose lip gloss (sorta) tube for scale

Replied in thread

@heinrichsgeist @katyswain @AustRealProg

I agree that while MMT presents as 'descriptive', to assume that human nature does not imbue descriptions with personally relevant meaning is to misunderstand human behavior.

It's actually quite simple to describe this without any judgement or subjective/political leanings. (It may be debated whether or not #systems philosophy is itself an ideology..)

The part where mmt shows taxes to be a dead-end in the financial flow, until spent back into the economy can be simplified/generalized further. Any system with flow will have the same set of logical parts that controls its physics. You need to know the volume that's moving, which includes both the velocity and size of the medium (like pipes & wires) it moves through the subsystems.

"Rich money" and "poor money" do not spend the same in an economy because of the important differences in their flows. Both taxes and profits inflate/increase the totals moving through the system, but to different areas. Profits flow up into the hands of the wealthy, and then 'trickle down' to the economy again (maybe, someday). It's very unidirectional, and most goes laterally to other wealthy nodes.

Taxing profits is *more* important than taxing incomes for this reason, otherwise, it's dead-end value for the system as a whole.