Scan Business provides a one stop shop for all your infrastructure needs - from workstations to servers, storage to networking, services to consultancy. Our team is expert in providing high performance datacentre solutions to support demanding applications such as high performance computing (HPC), virtual GPU (vGPU) and big data analytics.
We'll show you how to accelerate digital transformation in your workplace - whether you're self employed, or work in an SMB, public sector organisation or enterprise company. Our content focuses on getting the best out of existing technology and understanding how to lever cutting edge solutions in your everyday business processes to improve productivity, reduce bottlenecks and most importantly, drive competitive advantage.
Scan Business comprises specialist divisions, check out Scan AI, Scan Pro Graphics, Scan Cloud and Scan Digital Signage by clicking on the Channels tab above.
Hey i got a small network server with some 1tb hdds that sound not that good and i got 2 1tb intenso ssds How long will they last in the server I usually use 200-300 gb of them
Great video...Thanks! Happy to subscribe. I'd like to lay on a bit of praise and appreciation... I'm a senior with a 24/7 Tinnitus. I experience a "Cicada" type humming/ringing noise that resonates in my ears and head. It can make absorbing video audio/dialog difficult if the presenter doesn't annunciate well. Your presenter was very precise, articulate, and well paced. It made it very easy to hear and understand over my distraction. 🤗
Thanks a lot for your video. One question, can L4 be used for basic rendering tasks? On one of your tables this feature is crossed out for this particular GPU
Hi, there's nothing stopping you from using a L4 for rendering, it just isn't fast at it, so we wouldn't recommend buying one for that task. However, if you've got an L4 already there's no harm in sending some rendering it's way... - JG
Explanation is good😊.but doesn't explain about how much exact download and upload speed take per port as data transfer speed.its okays it might be plan speed÷8
I'm just super sad about all the rumors that they gonna push cumulus over onyx even more. None of my tiny customers wants it. We picked mellanox specifically for the unique mix of performance and focus on necessary features. I wish they'd use their vast resources to make this even better.
Thanks for the video. I work mainly with A100 and A10 for ML workloads, but I didn't understand what the purpose of the L series is. Thanks for shedding light on this 💡 I'm wondering why aws still offers K and M series instances at fairly high prices. Are there legacy systems that can't switch to modern hardware?
Glad to hear you found it useful. Thanks for the feedback. K and M GPUs are indeed very much legacy models, from 2012 and 2015 respectively, so will be pretty hopeless for ML/AI workloads. We can't really say why AWS is charging so much for instances like this, but all cloud providers are bound to have some older GPUs kicking around. Scan will be launching a new and improved cloud service in a few weeks, watch this space for the announcement. - JG
As informative as always. I think it's worth considering whether the GPUs are required for training (large batch sizes) or inferencing (small batch sizes). Nvidia's big boys are effective for training but too costly and inefficient for inferencing. Apart from large tech companies, most users are likely to deploy pre-trained models (and may be a bit of finetuning on small datasets) and therefore shouldn't overspend on the most expensive hardware (and lengthy waiting time).
Thanks for the thoughtful comment. These less demanding inferencing workloads are one of the major reasons why NVIDIA has the L4 and A2 GPUs. We'll try and make that point clearer in future videos. - JG
That's more CPU-limited, than GPU-limited, but you could probably run dozens of Minecraft servers on the sorts of servers these GPUs are normally installed in. - JG
What I would like is to understand the actual differences within the Xeon CPU oriented for Workstation and server. Both of them have the AVX-512 instruction feature so, apart from the primary and secondary cores and the capability to overclock ... Is there any other difference?
The biggest difference between Xeon Scalable (servers) and Xeon W (workstations) comes down to two main things. Firstly, Xeon W are available at a higher clock speeds (4.8 vs 4.1GHz), which helps speed up content creation applications, and conversely holds Xeon Scalable's back. Secondly, Xeon W have more PCIe lanes (112 vs 80) which helps when connecting lots of GPUs, SSDs etc, speeding up workloads such as data science and rendering. - JG
It's almost impossible to give a simple answer to this as server workloads vary so much, but.... in general I think it's fair to say that 4th gen EPYC still has a lead over 5th gen Xeon as they have: more cores, more PCIe lanes and higher memory bandwidth. AMD's architecture also appears to be more efficient than Intel's. Select (but not all models) of 5th (and 4th gen) Xeons do have workload specific accelerators that might give them an edge in some applications, but that's going to be down to ISV support. - JG
I'm afraid not, 5th gen Xeon Scalable are new CPUs for Socket 4677 for C740 boards. X99 boards are for Xeon E5 v3 Socket 2011 processors from 2014. - JG
Simple yet informative video. Thank you guys. Personally I see the 5th gen Xeon as a "refresh" to the 4th gen (same story goes for the Intel core 14th gen). Later this year intel will introduce CPUs with smaller transistors, which should be very cool (including SKUs with 288 cores!). The main reason why one would pick an Intel Xeon over and AMD EPYC is probably the accelerators - assuming you have a software that can leverage them, otherwise the higher core count AMD provides may be beneficial.
Thanks for the feedback. Releasing videos in 4K is something we're looking into, but we have to weigh up the time and storage required to shoot at a higher resolution than 1080p. - JG
Hello Scan IT Solutions, I recently purchased a Precision 5510 machine with an Intel(r) CPU E3-1505M v5 for my Python programming and web development needs. I'm eager to know more about how well it performs in these fields. Could you please provide a brief explanation of its capabilities and how efficiently it can handle programming tasks? Thank you in advance for your insights! Best regards
Hi Ali, the Xeon E3-1505M v5 CPU in your system has four cores and eight threads so should handle basic programming and web dev stuff ok. However, due to its age (2015 - Skylake architecture) it's going to struggle with more complex projects, i.e. compiling big applications or image and video work. For example, the current equivalent CPU, the Xeon E-2176M has six cores and twelve threads and runs on a more architecture (2018 - Skylake). - JG
Are there any Linux + Windows compilers that can take advantage of the new CPU instructions. Even if it means I compile to .O or .OBJ files and link to my current projects.
Hi, in short, yes. The main additional instruction set on EPYC 9004 is AVX-512, which is natively supported by multiple compilers. AMD Zen Software Studio contains a bunch of development tools, notably including AOCC compiler and libraries as well as a handy profiling tool - all accessible free of charge at www.amd.com/en/developer/zen-software-studio.html and fully optimised for AMD's own architectures. GCC 13 contains similar support, and if you’re tied to the Intel toolchain, that will also work. Happy compiling - JG
