yes ... especially if you want to execute quantum-circuits which use a lot of qubits.
why!? one approach of the simulation of quantum-computers rely on the so called "state vector" of the machine, and its memory-usage grows exponentially.
i've used the ibm quantum platform together with python/qiskit during my last project - which was something like: simulate quantum-networks on "real" quantum-computers...
ibm's support, introductions / documentation anbd usability of the platform is really great.
idk ... not comparable / much better than most of the quantum-computing hardware startups i know / looked at. of course, its easy if you have "deep pockets" like ibm does ... ;))
ok, back to the quantum platform:
it had a free-tier on the "old" quantum-platform - until july 2025: 10 minutes of compute on a set of machines - back then up to 127 qubits - per month ... no identification necessary / just an email-address.
sadly this "very generous" free-tier was killed of during the transition to the all new "quantum cloud platform" during spring/summer 2025 ...
i'm a veteran technology professional (25+ years) with experience in a variety of software-development, system-architecture, systems-administration, service-reliability-engineering and devops-/cloud-engineering (container / kubernetes) roles.
yes please, remove the crazy US of A from NATO already.
maybe then we are able to concentrate on 2 essential things here in europe:
* defend our countries
and they are not defended in west-asia or anywhere else on the planet where the USA decides to "causally" bomb yet another country with no meaningful aim or even grasp of the reality of the "problem" at hand.
* build weapons for the use of defense
not "weapons" which are only there to enrich their producers like in the USA.
remember: NATO was meant to defend against the soviet-union.
it dissolved in 1991, case closed.
since then NATO is more or a less just a "tupper-ware" party like sales-channel for overpriced crap from the USAs MIC.
just my 0.02€
ps. and always remember frank zappa, who said in 1981 (!) something like:
us-american politics is the entertainment-division of the MIC.
about 20 or 25 years ago i used whatever old hardware i could find in someones cellar or a junkyard together with 2 NICs and a floppy-disk drive / FDD based linux-distribution ...
it outgrew its original media - FDD - and is still active, as a router-focused distribution:
>I had somehow pigeonholed it as "for cloud to local or vice-versa", and never considered it for local transfer, like over my own LAN.
Huh. Exact same thoughts here. I never used rclone because I don't use AWS or similar, just lots of bare metal on and off my LAN that I can ssh to. I am quite comfortable with rsync's quirks and args, so not sure I'll quit using it, but maybe I'll try rclone next time I do a massive transfer. Similarly I've been shown that ddrescue can act as a full dd replacement, but I'm so used to dd I still tend to use it.
yes ... especially if you want to execute quantum-circuits which use a lot of qubits.
why!? one approach of the simulation of quantum-computers rely on the so called "state vector" of the machine, and its memory-usage grows exponentially.
for example qiskit AER
* https://qiskit.github.io/qiskit-aer/stubs/qiskit_aer.Stateve...
just as an example:
for 32 qubits, the simulator needs 64 GB RAM
=?> double the RAM for each additional qubit
so: for 36 qubits, the simulator needs 1 TB RAM
:)
so it gets pretty "costly" to do simulations rather quickly ...
just my 0.02€
reply