When you have heterogeneous SSDs, e.g. you mix PCIe 5.0, PCIe 4.0 and Thunderbolt interfaces, you can obtain a greater throughput by managing in software the distribution of data, than by using RAID0.
RAID0 works fine only when all the interfaces have the same speed.
If one SSD is twice faster than the other, in order to achieve maximal throughput, you must take care to place the data in such a way so that you will need to read twice more data from the twice faster SSD.
In general, you must distribute the data so that the amounts read from each SSD are proportional with the throughputs of the SSDs.
One could write a modified RAID0 device driver, which would use unequal stripes, with widths proportional with the SSD throughputs, but I am not aware of any such already existing RAID0 driver.
This would not work with HDDs, because the time to seek between 2 partitions of the same device would cause abysmal performance, but on SSDs this should work fine, even if some SSD controllers might still have a lower throughput when reading non-sequential pages.
It describes the culture of both companies at that time very clearly.
I think the difference between Apple and Microsoft is that Steve was a "product" CEO. He would bot only send that email, but in addition he would somehow oversee/sypervise those changes ALL the way until he was satisfied.
Meh, from the documentaries I have seen and reading about her, to me it seems that she is hungry for attention. She will do whatever she vmcan to get attention, and she has succeeded once more with this documentary. She is also getting your money.
I dont understand why dont people stop beating that death horse. She is a scanner and fraudster. She defrauded a lot of people and is in jail for that. Move on. Let her rot in the obscurity in she belongs to
I subscribe to a similar view: I admire things and ideas produced by "worse people" (people that have done worst things in my opinion) than this dude.
It's like the Iced Earth singer drama, or Kevin Spacey. I admire their work, I enjoy their performances, still I think their actions are deplorable and will be happy if they are brought to justice.
Ooooh my you made me feel old (I'm 45). We used to manually turn LPs backwards to do the same. Or sometimes press the PLAY and << buttons at the same time in tape players to try to achieve the same.
Principal AI/ML Engineer and hands-on engineering leader (CTO, VP, Head of Engineering), 20+ years in software, PhD in Computer Science. The last 8 years have been remote fintech roles where I stayed in the code while building and running teams.
Currently at Aloi.tech (AI tax assistance): LLM pipeline for invoice analysis with RAG, fine-tuning of open-source and closed LLMs, and a custom OLAP framework for GB-scale invoicing data. Before that, CTO at Bax Blockchain Services: architected a B2B crypto exchange from scratch (event-driven microservices, RabbitMQ, AWS ECS, Terraform, multi-environment IaC), owned security (WAF, GuardDuty, Inspector, WireGuard), operated a Liquid Network node and hot wallet, and led a 20+ person engineering team. At PrimeTrust, VP of AI/ML & Lead Architect: built the AI/ML team for fraud detection (graph analysis on AWS Neptune, K-NN anomaly detection), designed a Cassandra ledger that took transaction processing from hundreds to tens of thousands per second, and drove the monolith-to-microservices migration. At Paystand, Head of Engineering & Product: led SOC 2 and PCI-DSS Level 1 compliance, cut AWS costs 30%, and grew the Mexico engineering office from 1 to 15. Earlier: Head of Engineering & Data Science at Kueski (ML credit risk models in production, ISO 27001), Ooyala, and academic research.
Looking for: Principal/Staff AI/ML Engineer, ML/LLM infrastructure, CTO / VP / Head of Engineering at an early-to-growth-stage startup, fractional CTO or technical advisor, or founding engineer on a small technical team. Fintech, legaltech, and data-heavy products preferred but not required.
More or less. I think the use of CPU specific instructions can make an compiled program different than "the rest". Although nowadays some compilers are clever enough to do better than manual optimization.
Michael Abrash wrote about such optimizations on the late 90's regarding Pentium versus its predecessors, and then everyone started using Python, Ruby, whatever for full stack applications, beyond plain OS scripting tasks.
reply