Nvidia shows signs in [2020]

[Deleted member 2197]

VMWARE EMBRACES NVIDIA GPUS, DPUS TO DRIVE ENTERPRISE AI
September 30, 2020

VMware chief executive officer Pat Gelsinger says organizations want AI for such tasks as video analytics and real-time streaming for fraud detection, but added that “as exciting as these next-generation apps are, they’re beyond the reach for mainstream organizations. In fact, enterprise AI adoption is stuck at just 10 to 15 percent.”
...
Integrating the Nvidia NGC suite into its key hybrid cloud offerings represents a change for VMware. The company in its journey from datacenter virtualization pioneer to hybrid cloud solutions provider has been primarily X86 CPU-focused. However, GPUs – which started off as graphics chips for devices and under Nvidia’s relentless drive have become key accelerators in the datacenter – are becoming foundational tools for AI and other emerging workloads.
...
The key shift in the architecture is from basing it on core CPUs to SmartNICs, based on Nvidia’s Mellanox BlueField-2 data processing unit (DPU). With Project Monterey, VMware can run its ESXi hypervisor, a move that required porting ESXi to the Arm architecture. Nvidia’s SmartNICs are based on the Arm architecture, which isn’t surprising given Nvidia’s past use of the architecture and the fact that Nvidia is now in the process of buying Arm for $40 billion. In the new architecture, there are two ESXi instances for each physical server – one on the primarily x86 processors and the other on the SmartNIC – and they can run separately or together in a single logical instance. Storage and network services also run on the SmartNIC, which improve the performance of both while reducing pressure on the CPU. The SmartNIC ESXi will manage the x86 ESXi.

The new highly disaggregated architecture also exposes the hardware accelerators – like GPUs and FPGAs – to all hosts in a any cluster to allow applications in the cluster to leverage the accelerators both ESXi and bare-metal environments.
...
Nvidia developed the BlueField-2 DPU for Project Monterey, Huang said, adding that it’s “built on the Mellanox state-of-the-art, well-known high performance NICs. The BlueField DPU Is going to essentially take the operating system of the datacenter — networking, storage, security, virtualization functionality — and offload it onto this new processor. This new processor is going to be essentially the datacenter infrastructure on a chip. Datacenters are going to be much more performant result of this.”

...
This leads us into an era of disaggregation where, rather than deploy intact systems, we aim to deploy smaller, malleable building blocks that are disaggregated across a fabric and must be composed to realize the intent of the user or application. The provisioning of engines to drive workloads is completely API-driven and can be specified as part of the Kubernetes manifest if using VCF with Tanzu. We call this intent-based computing.”

Vendors like Dell and HPE have been talking about composability for the past several years, but the concept can be seen in mainframes and old Unix-based systems. With x86 systems, there has been “coarse-grained” composability with technologies like VMware’s Software-Defined Data Center (SDDC), which created software-defined infrastructure out of intact servers or storage systems. Project Monterey will deliver more fine-grained composability, including extending disaggregation to the hypervisor by making most of the general-purpose compute available via the SmartNICs, he wrote.

https://www.nextplatform.com/2020/09/30/vmware-embraces-nvidia-gpus-to-drive-enterprise-ai/

 

[Deleted member 2197]

NVIDIA Introduces BlueField DPUs to Bring Performance to Every Data Center

NVIDIA today announced a new kind of processor DPUs, data processing units—supported by DOCA, a novel data-center-infrastructure-on-a-chip architecture that enables breakthrough networking, storage and security performance.
...
Optimized to offload critical networking, storage and security tasks from CPUs, BlueField-2 DPUs enable organizations to transform their IT infrastructure into state-of-the-art data centers that are accelerated, fully programmable and armed with "zero-trust" security features to prevent data breaches and cyberattacks.A single BlueField-2 DPU can deliver the same data center services that could consume up to 125 CPU cores. This frees up valuable CPU cores to run a wide range of other enterprise applications.

Widespread Adoption of NVIDIA DPUs
Leading server manufacturers worldwide—including ASUS, Atos, Dell Technologies, Fujitsu, GIGABYTE, H3C, Inspur, Lenovo, Quanta/QCT and Supermicro—have plans to integrate NVIDIA DPUs into their enterprise server offerings.

These commitments from system providers are complemented by extensive support from software infrastructure partners, including:

• VMware announced substantial work underway with NVIDIA as part of its recently announced Project Monterey initiative to support BlueField-2 DPUs with VMware Cloud Foundation.
• Red Hat plans to offer support for BlueField-2 DPUs with Red Hat Enterprise Linux and Red Hat OpenShift, components of Red Hat's open hybrid cloud portfolio, which is used by 95 percent of the Fortune 500.
• Canonical announced support of BlueField-2 DPUs and DOCA in its Ubuntu Linux platform, the most popular operating system among public clouds.
• Check Point Software Technologies, a leading cybersecurity provider, is integrating BlueField-2 DPUs into its technologies, which more than 100,000 organizations worldwide use to protect themselves from cyberattacks.

NVIDIA's current DPU lineup includes two PCIe products:

• The NVIDIA BlueField-2 DPU, which features all of the capabilities of the NVIDIA Mellanox ConnectX -6 Dx SmartNIC combined with powerful Arm cores. Fully programmable, it delivers data transfer rates of 200 gigabits per second and accelerates key data center security, networking and storage tasks, including isolation, root trust, key management, RDMA/RoCE, GPUDirect, elastic block storage, data compression and more.
• The NVIDIA BlueField-2X DPU, which includes all the key features of a BlueField-2 DPU enhanced with an NVIDIA Ampere GPU's AI capabilities that can be applied to data center security, networking and storage tasks. Drawing from NVIDIA's third-generation Tensor Cores, it is able to use AI for real-time security analytics, including identifying abnormal traffic, which could indicate theft of confidential data, encrypted traffic analytics at line rate, host introspection to identify malicious activity, and dynamic security orchestration and automated response.

https://www.guru3d.com/news-story/n...o-bring-performance-to-every-data-center.html

 

[Deleted member 2197]

Edge AI gets edgier

For the world of edge AI, Nvidia announced a new entry-level developer kit, the Jetson Nano 2GB. This kit will retail at $59 and is designed for teaching and learning AI. It will be available at the end of this month.

“In the first five years of the Jetson journey, we started from zero and went up to 200,000 developers,” he said. “In the last 18 months, we’ve added another 500,000 developers, and we also observed that the activity of these developers has increased 10 times… I think we are at a seminal point right now where every software engineer in the world wants to reskill themselves to become an AI engineer. Every university student in engineering, whether it’s electrical engineering or computer science wants to jump in re-skill themselves with AI. And we’re also starting to see high school students, especially in the STEM area, wanting to learn AI.”

Futuristic communication

Omniverse is a 3D simulation and collaboration platform that allows remote teams to collaborate on projects that rely on 3D graphics, including architecture, engineering, animation and more. The open Beta follows a year-long early access program in which 40 companies provided feedback to Nvidia.

Omniverse uses Pixar’s widely adopted Universal Scene Description (USD) format for interchange between 3D applications, plus Nvidia real-time photorealistic rendering, physics, materials and interactive workflows between industry-leading 3D software products.

Nvidia also announced a new suite of cloud-based AI technologies for video call providers. Called Nvidia Maxine, the suite uses AI to correct the gaze of participants so it looks like they are always looking into the camera, creates super-resolution video from lower quality streams, artificially adjusts lighting, and more.

“A lot of us as we’re on these calls all day have multiple windows open,” said Richard Kerris, general manager of media and entertainment at Nvidia. “We’re looking at different things, and not really making the eye contact that you want to make with the person that you’re talking to. Using AI we can actually reconstruct that face and ensure that the eye contact is taking place so that you have a more personalized experience.”

Maxine can also reduce the bandwidth needed for video calls. Instead of simply streaming the whole video, Maxine’s AI can determine facial movements of the person speaking, and send only that data over the internet. It then reanimates the person’s face from the data at the other end. This tech can reduce video bandwidth to a tenth of what’s required for H.264 streaming today.

Other features for video call providers include use of Jarvis, Nvidia’s conversational AI, to translate between languages in real-time, so you can video conference with a person speaking another language.

https://www.eetimes.com/nvidia-presents-the-dpu-a-new-type-of-data-center-processor/#

 

[Deleted member 2197]

Nvidia and EuroHPC Team for Four Supercomputers, Including Massive ‘Leonardo’ System

The EuroHPC Joint Undertaking (JU) serves as Europe’s concerted supercomputing play, currently comprising 32 member states and billions of euros in funding. In June 2019, EuroHPC selected eight supercomputing centers across the European Union to serve as hosts for EuroHPC’s first eight systems.

Today, Nvidia is announcing that it will be powering at least four of those eight systems: Leonardo, hosted by CINECA in Italy; Meluxina, hosted by LuxConnect in Luxembourg; EURO IT4I, hosted by the IT4Innovations National Supercomputing Center in the Czech Republic; and Vega, hosted by IZUM in Slovenia. All four systems will utilize Nvidia Ampere GPUs and Nvidia HDR InfiniBand networking, and Atos will serve as the prime contractor for all but EURO IT4I (which will be built by HPE).
...
CINECA’s Leonardo system, however, is the belle of the ball: a computing juggernaut expected to serve up over 200 peak petaflops and cost up to €240 million (half from EuroHPC and half from the Italian Ministry of University and Research). Leonardo is one of three planned pre-exascale systems announced by the EuroHPC JU.

Leonardo will contain a whopping ~14,000 Nvidia A100 GPUs distributed among ~3,500 Atos Sequana nodes, each equipped with four A100s and a single Intel CPU slotted on Nvidia’s HGX baseboard. The water-cooled nodes will use Nvidia Mellanox HDR 200Gb/s InfiniBand networking, which CINECA says is boosted by smart in-network computing acceleration that enables low latency and high data throughput. Per Nvidia, this acceleration yields a 7X speedup in reduction operations, a 2X speedup in MPI performance and a 10X speedup in data movement among remote GPUs.
...
Nvidia anticipates that Leonardo will deliver 10 exaflops of half-precision AI performance (with the A100’s structural sparsity feature enabled), and everyone involved is eager to crown the new system as the world’s fastest AI supercomputer. While results from HPL-AI and MLPerf benchmarking will need to wait for a real-world system, Nvidia expressed confidence that Leonardo will perform well and stressed the necessity of strong AI performance in modern supercomputing.

Subject to the timely construction of the datacenter, Leonardo is scheduled to begin deployment in 2021 and become fully operational sometime in 2022. Once operational, CINECA will use Leonardo for a wide variety of research activities.

https://www.hpcwire.com/2020/10/15/...rcomputers-including-massive-leonardo-system/

 

[Deleted member 2197]

Hyundai, Nvidia expand ties for connected cars
November 10, 2020

Starting from 2022, the automotive group said it will apply the Nvidia Drive connected car platform to all upcoming vehicles of its three car brands ― Hyundai Motor, Kia Motors and its luxury car brand Genesis.

Under the agreement, every new Hyundai, Kia and Genesis vehicle will come standard with the connected car platform. Nvidia Drive is a computer platform designed for cars, aimed at providing autonomous driver assistance functionality powered by deep learning technology.
...
To meet customer demand, Hyundai Motor Group said the next-generation connected car system will be able to process enormous amounts of data quickly, offer a flawless ...
connecting experience between the vehicle and surrounding infrastructure, be able to understand drivers' intentions and health condition and provide upgraded safety features.
...
"For Hyundai Motor Group, standardizing the high-performance, energy-efficient Nvidia Drive platform for its future models allows for a seamless and continuously enhanced in-vehicle AI user experience. Nvidia Drive includes a hardware and software stack, enabling Hyundai, Kia and Genesis IVI systems to feature audio, video, navigation, connectivity and AI-enhanced connected car services," Nvidia said.

Hyundai has been cooperating with Nvidia since 2015. The automotive group has used Nvidia's driving platform in its Genesis GV80 SUV and G80 sedan, which were launched this year.

https://www.koreatimes.co.kr/www/tech/2020/11/419_299084.html

 

[Deleted member 2197]

Nvidia sales and earnings hit records as videogame and server chips surge

Nvidia Corp.’s quarterly results blew past estimates to record highs Wednesday, with sales surpassing $4 billion for the first time as the chip maker rolled out new gaming cards and data-center demand stayed hot.
...
Nvidia introduced new gaming chips in the quarter, based on a new design that was rolled out in its server chips the previous quarter, when data-center sales outperformed gaming sales for the first time ever. The rollout of new gaming cards in September flipped that: Gaming sales surged 37% to a record $2.27 billion, while data-center sales — up 8% from the previous quarter — soared 162% to $1.9 billion from the year-ago period, also a record high.
...
For the fourth quarter, Nvidia looks to possibly set even more records, forecasting revenue of $4.7 billion to $4.9 billion, while analysts had forecast revenue of $4.4 billion on average.
Given the third quarter’s results, Nvidia forecast that segment sales would travel in opposite directions to what Wall Street analysts were expecting.

“We expect gaming to be up sequentially in what is typically a seasonally down quarter as we continue to ramp up our new RTX 30 series products,” said Colette Kress, Nvidia’s chief financial officer, on the analyst call. “We expect data center to be down slightly versus Q3.”

https://www.marketwatch.com/story/n...-videogame-and-server-chips-surge-11605735691

 

[Erinyes]

Nvidia sales and earnings hit records as videogame and server chips surge

https://www.marketwatch.com/story/n...-videogame-and-server-chips-surge-11605735691

So Nvidia had a record quarter with and record gaming revenue and they're also projecting similar revenues for the upcoming quarter. So where are the Ampere cards in the market? Is it just simply a matter of extreme demand outstripping supply? Several retailers and AIBs have said otherwise but then where is the revenue coming from? A significant number of lower end Turing GPUs and mobile GPUs perhaps?

 

[DegustatoR]

Is it just simply a matter of extreme demand outstripping supply?

Ding ding ding

 

[Kaotik]

Ding ding ding

Extreme demand is there, but it is by no means normal that a major retailer serving several countries (all of the nordics, germany, poland, austria) had received two months after the sales began only 901 cards of over 9300 they had ordered from manufacturers (3080, the others numbers aren't really much prettier than that)
https://www.proshop.de/RTX-30series-overview

It's not only extreme demand, lack of proper supply is a big portion of it.

 

[DegustatoR]

Extreme demand is there, but it is by no means normal

I agree, extreme demand is by no means normal.
There are no other explanation left.
There is no revenue fall off in the last quarter results. It has actually increased substantially in both Q/Q and Y/Y. And this cannot be attributed to anything but Ampere launch.
AMD is experiencing the exact same issues with Zen3 and N21 which means that there are no problems with Samsung production or G6X availability.
Sorry but all these theories are pure bs. It's just extreme demand of this weird year.

 

[Deleted member 2197]

Video card inventories were sufficient to meet and sustain a 37% quarterly revenue increase in booked video card sales, so Nvidia exceeded it's inventory objectives for the quarter.
Sounds like Nvidia had a lot more small to medium sized e-retailers selling cards than in the past (perhaps due to Covid-19). Only very extreme demand could create conditions where low inventory levels and record breaking revenues are at loggerheads. Good news is adjusted 4th quarter forecast indicates inventory levels on pace to shortly meet extreme demand.

 

[Kaotik]

I agree, extreme demand is by no means normal.
There are no other explanation left.

Don't cut the rest out, as you very well know your quote doesn't represent what I said.

There is no revenue fall off in the last quarter results. It has actually increased substantially in both Q/Q and Y/Y. And this cannot be attributed to anything but Ampere launch.
AMD is experiencing the exact same issues with Zen3 and N21 which means that there are no problems with Samsung production or G6X availability.
Sorry but all these theories are pure bs. It's just extreme demand of this weird year.

Extreme demand doesn't explain lack of supply no matter how many times you try to repeat it.
Shipping 900 cards in 2 months to a major retailer, who has ordered over 9300 cards is lack of supply no matter how the demand is.

AMDs problems can be same, or they can be different, it all depends on how the supply looks. We currently don't know solid numbers on AMDs Zen 3 supply (I do however know, that at least biggest finnish player in the field had more of them than RTX 3080's on launch). RX 6800 -supply is limited like RTX 30 for reference, but it might correct itself* in a week if rumors are correct (they're talking shipments around 6-7 times bigger than RTX 30 for AIB models)

*correcting itself doesn't mean they still wouldn't sell out

 

[DegustatoR]

Don't cut the rest out, as you very well know your quote doesn't represent what I said.

I will cut out whatever I want when I see someone spreading FUD.

Extreme demand doesn't explain lack of supply no matter how many times you try to repeat it.

Extreme demand explains the lack of availability, and you know nothing about the supply beyond some spreadsheet from a couple of regional retailers.
The only real way of assessing overall supply volume is the revenue results since this is money which the seller gets from shipping the product.
You have them now. I see nothing to discuss here further.

AMDs problems can be same, or they can be different, it all depends on how the supply looks. We currently don't know solid numbers on AMDs Zen 3 supply

Oh, would you look at that, suddenly we don't know this for AMD but know for Nvidia.
Stop being a tool.

Sorry for breaking the silence, will go back into it again.

 

[Kaotik]

I will cut out whatever I want when I see someone spreading FUD.


Extreme demand explains the lack of availability, and you know nothing about the supply beyond some spreadsheet from a couple of regional retailers.
The only real way of assessing overall supply volume is the revenue results since this is money which the seller gets from shipping the product.
You have them now. I see nothing to discuss here further.


Oh, would you look at that, suddenly we don't know this for AMD but know for Nvidia.
Stop being a tool.

Sorry for breaking the silence, will go back into it again.

We know of NVIDIA because we have major retailer who has given us solid data on supply.
We do not know of AMD because we do not have any retailer who would have given us any data on supply.

I have provided data to back my claims, you haven't.

 

[DegustatoR]

We know of NVIDIA because we have major retailer who has given us solid data on supply.
We do not know of AMD because we do not have any retailer who would have given us any data on supply.

I have provided data to back my claims, you haven't.

Sigh.

https://www.scan.co.uk/shops/amd/ryzen-5000-faqs

 

[Kaotik]

Sigh.

https://www.scan.co.uk/shops/amd/ryzen-5000-faqs

Thank you, now you've provided data to give your claims on AMD CPU supply something to stand on. (And I really wasn't aware of that even though I've been trying to follow these supply issues)
Looks like Ryzen 5000 is suffering from similar supply issues unless those "TBC" amounts turn out to be huge, since Scan probably has ordered a lot more than they've gotten. Now we just need to wait for RX6800 data and how the AIB-launch next week affects that.

 

[Erinyes]

I agree, extreme demand is by no means normal.
There are no other explanation left.
There is no revenue fall off in the last quarter results. It has actually increased substantially in both Q/Q and Y/Y. And this cannot be attributed to anything but Ampere launch.
AMD is experiencing the exact same issues with Zen3 and N21 which means that there are no problems with Samsung production or G6X availability.
Sorry but all these theories are pure bs. It's just extreme demand of this weird year.

Misattributing quotes still provides no data to back up your claims. Certainly there is some Ampere revenue, no one's disputing that. But there is also ample evidence of the lack of supply. Can you state for a fact that Turing revenue did not in fact go up? No, you can't. Ampere offers nothing below the $500 price point where allegedly the most GPU sales happen, nor laptops, which also constitute a large part of GPU sales today.

 

[trinibwoy]

Looks like Ryzen 5000 is suffering from similar supply issues unless those "TBC" amounts turn out to be huge, since Scan probably has ordered a lot more than they've gotten.

We’re sorta pissing into the wind with supply speculation. Nobody knows what’s happening with supply except AMD and Nvidia.

I randomly scored a 5950x at Microcenter the other day. They had just put about 10 boxes on the shelf a few mins before I walked in. They also never updated the website to show the store had stock. I wouldn’t place too much faith in numbers from any retailer.

Now that we know Nvidia is still making money hand over fist the answer is that they must be selling a boat load of Turings. I’m sure that’s true but it says nothing about Ampere’s numbers. Why is it so hard for people to accept that we just don’t know.

 

[Deleted member 2197]

Now that we know Nvidia is still making money hand over fist the answer is that they must be selling a boat load of Turings. I’m sure that’s true but it says nothing about Ampere’s numbers. Why is it so hard for people to accept that we just don’t know.

Talk about pissing in the wind. The only valid information is what retailers mention as unprecedented demand for Ampere. Prices for Turing products on Newegg and Amazon really have not dropped that much to justify purchases over Ampere products, and 2nd hand sales don't contribute to Nvidia's bottom line.

 

[Picao84]

I grabbed an Asus 3070 Dual OC from ebuyer, a lesser known retailer that is mostly used by small and medium companies to buy equipment of. They seem to be getting some stock everyday from a month after RTX3080 release. They don't do queues though, so potential buyers better keep refreshing the page. That's how I got mine.

Additionally, it seems that German shops have frequent supply, judging by several OCUK customers that cancelled their order and got them from Germany.