Tag: GPUs

Lenovo unveils first all-AMD AI ‘supercomputer’ flanked with up to 1.5TB of HBM memory and promises drop-in support for the future AMD EPYC CPU — new ThinkSystem has dual EPYC CPUs and 8 Instinct MI300X GPUs

Post author By lisa nichols
Post date April 29, 2024
No Comments on Lenovo unveils first all-AMD AI ‘supercomputer’ flanked with up to 1.5TB of HBM memory and promises drop-in support for the future AMD EPYC CPU — new ThinkSystem has dual EPYC CPUs and 8 Instinct MI300X GPUs

[ad_1]

Lenovo has taken the wraps off its ThinkSystem SR685a V3 server, which it says is an optimal solution for both enterprise private on-prem AI as well as for public AI cloud service providers.

Crafted in tandem with AMD, the server has been specifically engineered to handle the demanding compute needs associated with GenAI and Large Language Models. With fast acceleration, large memory capacity and I/O bandwidth, the new powerhouse can manage sizable data sets used in industries from healthcare and energy to financial services and climate science.

The AI supercomputer is powered by two top-tier 4th Generation AMD EPYC processors, each delivering up to 400W of output, coupled with eight AMD Instinct MI300X GPUs. Fully interconnected with AMD Infinity Fabric, it provides 1.5TB of high-bandwidth (HBM3) memory capacity and up to 1TB/s peak aggregate theoretical GPU I/O bandwidth performance.

Up to 16 NVMe SSDs

ThinkSystem SR685a V3 supports up to 3TB RAM, made up of 24 DDR5 DIMMs, and as an alternative to the eight AMD Instinct MI300X GPUs with Infinity Fabric interconnects at 896GB/s, it can handle eight Nvidia H100/H200/B100 GPUs with NVLink interconnects at 900GB/s.

The 8U2S rack server allows for up to 10x PCIe Gen5 x16 FHHL adapters (eight at the front connected to the PCIe switch for GPU connectivity, and two at the rear linked directly to the main CPU). It supports up to sixteen 2.5-inch hot-swappable NVMe SSDs, as well as up to two M.2 drives for boot, with RAID support. In terms of power, eight hot-swappable PSUs allow for full N+N redundancy, while things are kept at an acceptable temperature via air cooling with an N+1 hot-swappable fan.

Management is handled by an XClarity Controller2 (XCC2), that offers high-level service-processor control as well as monitoring and alerting functions.

The ThinkSystem SR685a V3 is compatible with a variety of operating systems including RHEL, Ubuntu, Alma Linux, Rocky Linux, and ESXi.

“Together with Lenovo, we are accelerating AI transformation for the enterprise in multiple manners – helping guide customers through this monumental shift in computing, while providing robust end-to-end solutions at scale,” said Forrest Norrod, executive vice president and general manager, Data Center Solutions Group, AMD. “We have a proven track record of bringing amazing solutions, jointly, to the market and are excited to add AMD Instinct MI300X to Lenovo’s portfolio, helping enterprises to rapidly adopt AI.”

With applications spanning multiple sectors, this new system can accelerate most compute demanding AI workloads, and Lenovo singles out its benefits to the world of financial services in particular, where it can help detect and prevent fraud, and streamline KYC (know your customer) initiatives.

The Lenovo ThinkSystem SR685a V3 server can be ordered now, with pricing available on request.

Intel unveils flurry of new Arc GPUs — however serious graphics users will have to wait for more powerful models, as these focus on a completely different and more lucrative market

[ad_1]

Two years after the debut of its Arc Alchemist GPUs, Intel is launching six new Arc products, but these are designed for edge/embedded systems.

These edge systems, which process data near the source to reduce latency and bandwidth use, are becoming increasingly essential in areas such as IoT, autonomous vehicles, and AI applications.

As Intel says, “AI at the edge is exploding with new use cases and workloads being developed daily. These AI workloads often require a high degree of parallel processing and memory bandwidth for peak performance, dedicated hardware, optimized architecture for compute efficiency, and reduced latency with faster results for real-time processing. A discrete GPU may be the ideal solution for edge AI use cases requiring high performance and complex model support.”

Six SKUs

The new Arc on edge GPUs are built on Intel’s highly scalable Intel Xe-core architecture and support AI acceleration, visual computing and media processing. Using the OpenVINO toolkit developers can deploy AI models across Intel hardware.

The Arc on edge offerings have a number of benefits, including reduced latency, improved bandwidth efficiency and better privacy and security.

For high performance and to handle heavy AI workloads and expansive use cases such as facial recognition and generative conversational speech, there’s the 7XXE. For immersive visual experiences and enhanced AI inferencing capabilities, there’s the 5XXE, and for low power and small form factor requirements, Intel has the 3XXE.

There are six SKUs available – the A310E and A3503 with 6 Xe-cores, the A370E and A380E with 8 cores, and the A580E and A750E with 28 cores. The A310E, A3503 and A370E have 4GB of GDDDR6 memory with 112GB/s memory bandwidth. The A380E has 6GB with 186GB/s bandwidth, while the A580E and A750E’s memory and memory bandwidth are unknown for now. Intel says only that it is “in planning”. There’s also no launch date for those two either, just TBD. The other four SKUs will be available this month.

Intel Arc GPUs are built to be paired with Intel Core processors, from 10th Gen upwards and Intel Xeon W-3400 and W-2400. A number of products featuring the new Arc GPUs are set to be released in the coming months from Intel partners including from ADLINK, Advantech, Asus, Matrox and Sparkle.

Arc on Edge

(Image credit: Intel)

Nvidia’s RTX 5080 and 5090 GPUs might arrive later in 2024 — and it’s honestly way too soon

Post author By lisa nichols
Post date April 13, 2024
No Comments on Nvidia’s RTX 5080 and 5090 GPUs might arrive later in 2024 — and it’s honestly way too soon

[ad_1]

Nvidia’s next generation of graphics cards, reportedly known as Nvidia Blackwell, has been the subject of plenty of rumors since at least 2023.

The most recent one seems to confirm that the RTX 5080 and RTX 5090 will be launching in Q4 2024, according to Twitter leaker Kopite7kimi and reported on by PC Gamer. While finally getting a hold of what could easily be the best graphics cards sounds like great news to some, finally getting a sign of the cards’ imminent release, they’re honestly coming out way too soon.

There have been reports that the Nvidia RTX 5080 would have better ray tracing capabilities and pricing than the 4080 and that the RTX 5090 would be up to 70% faster than the 4090, which sounds great on paper. But when you consider how already powerful the 4000 series of cards are, it seems a bit pointless. It’s wonderful that ray tracing, speed, and performance might be improved, but to what benefit?

We aren’t even taking full advantage of current gen cards, so what’s the value of buying a Nvidia 5000-series GPU later this year?

PC games haven’t even reached the point that they truly challenge the mid-range and high-end 3000-series of graphics cards (hell, an argument can be made that they can’t fully challenge the RTX 2070 Super in ways that truly matter), let alone the current gen. One could argue that the only reason why the current gen was so necessary was thanks to the crypto mining craze completely ruining the 3000-series market and making it nigh impossible to get your hands on one for a remotely affordable price.

And right now, the 4000-series is quite excellent performance-wise, as it’s able to handle ray tracing and other advanced effects and tools like no other. The RTX 4090 in particular is nearly perfect in what it can offer, and the lower cards are still absolute performance beasts. This isn’t even mentioning the Super series refreshes, which added even more oomph to your favorite graphics cards while keeping the prices either the same or lowering them.

There’s also the fact that some cards, like the RTX 4070, didn’t sell nearly as well as Nvidia wanted and in fact were rumored to be a “disaster.” While that doesn’t reflect the sales numbers for the rest of the graphics cards, it’s not a good look to see the base versions of your mid-range cards doing poorly. And while the RTX 4080 Super seems to be out of stock in many retailers, that could just as well be due to low stock in the first place.

With all these issues cropping up, releasing RTX 5080 and 5090 doesn’t seem to be such a smart move on Nvidia’s part. Though the Q4 2024 launch date is still just a rumor, it would be wise for the tech giant to release these cards in 2025 at the earliest to give its 4000-series some proper breathing room.

[ad_2]

Source Article Link

Tags arrive, GPUs, honestly, NVIDIAs, RTX

Featured

Intel powers up XeSS upscaling to deliver faster frame rates for Arc GPUs – up to 30% quicker – and better image quality

Post author By lisa nichols
Post date April 5, 2024
No Comments on Intel powers up XeSS upscaling to deliver faster frame rates for Arc GPUs – up to 30% quicker – and better image quality

[ad_1]

Intel has unleashed a new version of its XeSS upscaling tech that rivals Nvidia DLSS and AMD FSR for speeding up frame rates in PC games.

Intel XeSS 1.3 is fresh on the scene, although note that it’s still in preview testing right now, with Team Blue claiming some decent performance boosts compared to XeSS 1.1 or 1.2.

On a PC with an Arc A750 GPU, Intel discovered a particularly impressive jump in frames per second (fps) with Diablo 4 at 1440p resolution (with high details and ray tracing on). Without XeSS, the action RPG ran at 37 fps, but with XeSS that increased to 71 fps – and with the new XeSS 1.3 the game hit 91 fps. That’s a performance gain of almost 30% over the previous incarnation of XeSS.

Other games have more modest, but still worthwhile, improvements, in the order of around 5 extra fps, a boost of around 10%.

Intel also demonstrated integrated graphics with the Arc GPU in an Intel Core Ultra processor, running games at 1080p with medium details. On average, Team Blue found XeSS 1.3 delivered 8% more fps compared to previous versions.

Not just faster, but better-looking upscaling too

XeSS 1.3 isn’t just about faster frame rates for your GPU, though, as Intel has been fine-tuning the image quality as well. The company tells us that version 1.3 of its upscaling tech benefits from an updated model that provides more detail in the image, better anti-aliasing (smoothing of jaggies) and less in the way of ghosting.

All in all, it seems like a sizeable stride forward for XeSS, but Intel will need to keep pushing with its upscaling technology – because AMD and Nvidia most definitely aren’t standing still.

Indeed, Nvidia just released DLSS 3.7 which offers similar improvements to what we see here from Intel (less ghosting, better temporal stability). And AMD may have even bigger moves underway – possibly finally bringing AI into the mix with FSR. (Note that both Intel and Nvidia already use hardware-accelerated AI cores to pep up their respective upscaling technologies).

Via Neowin

[ad_2]

Source Article Link

Tags Arc, deliver, faster, Frame, GPUs, image, Intel, powers, Quality, quicker, rates, upscaling, XeSS

Featured

The Nvidia RTX 4060, 4060 Ti, and 4070 GPUs could get refreshes soon

Post author By lisa nichols
Post date March 22, 2024
No Comments on The Nvidia RTX 4060, 4060 Ti, and 4070 GPUs could get refreshes soon

[ad_1]

Nvidia has been slowly launching its Super refreshes of the RTX 4000-series of graphics cards, updated versions with much-improved performance. And, according to a new rumor, the mid-range models might also be getting its own refresh.

There has been an alleged leak of plans for refreshed versions of RTX 4060, 4060 Ti, and 4070 graphics cards, which was revealed by Twitter/X user Zed__Wang and reported on by PCGamesN. Thanks to the table, we know the three brand-new dies for the cards as well as the release windows: March 2024 for the RTX 4070 and April 2024 for both the RTX 4060 and 4060 Ti.

table of rtx gpu specs — (Image credit: Zed__Wang)

While details for these newer versions are sparse, we normally could assume that a larger die comes with superior memory bandwidth, VRAM capacity, and performance. However, according to the same leaker, this is not the case.

When asked about the spec difference between the original and the refreshed GPUs, they responded “only change the chip; the spec keeps the same.” It seems like, at most, we might see a few additional CUDA, RT, and Tensor cores but that’s all.

These mid-range refreshes could be a dud

If this leak is to be believed, then this refresh sounds like a whole lot of nothing. If the broader specs aren’t even being upgraded, then what’s even the point of releasing them in the first place? The only way it would make sense is if the new cards were cheaper than the basic versions.

One good thing about the Super series that have previously launched is that it’s already been lowering the prices of AMD’s RDNA 3 graphics cards, going by the dropping prices of the AMD Radeon RX 7700 XT and the RX 7900 XT once the RTX 4080 Super, 4070 Ti Super, and RTX 4070 Super hit retailers. So if these newer mid-range cards end up launching, then that could mean the matching AMD cards could also see a price decrease.

It’s rather sad that these are the only positives of getting RTX 4060, 4060 Ti, and 4070 refreshes if the rumors are actually true. Here’s hoping that if Nvidia does decide to release them, they’ll be worth investing in.

[ad_2]

Source Article Link

Tags GPUs, NVIDIA, refreshes, RTX

Entertainment

NVIDIA says its Blackwell GPUs are the world’s most powerful chips

Post author By lisa nichols
Post date March 19, 2024
No Comments on NVIDIA says its Blackwell GPUs are the world’s most powerful chips

[ad_1]

NVIDIA’s H100 chips are used by nearly every AI company in the world to train large language models hooked into services like ChatGPT. It’s been great for business. Now, the company is ready to make those chips look terrible, announcing a next-generation platform called Blackwell.

Named for David Harold Blackwell, a mathematician who specialized in game theory and statistics, NVIDIA claims Blackwell is the world’s most powerful chip, reaching speeds of 20 petaflops compared to just 4 petaflops the H100 provided. Yeah, throw it in the trash. You need new chips.

And if you didn’t know how powerful NVIDIA is, its press release for this new platform includes quotes from the CEOs of OpenAI, Microsoft, Alphabet, Meta and Tesla — yes, all CEOs you probably know the names of.

— Mat Smith

The biggest stories you might have missed

You can get these reports delivered daily direct to your inbox. Subscribe right here!

The tournament is postponed until further notice.

Yeah, this is bad. Respawn, the EA-owned studio behind Apex Legends, has postponed the North American Finals tournament after hackers broke into matches and equipped players with cheats. Footage of the hacks on Twitch show players being able to see their opponent’s location through walls, while notable player (and one of the best) ImperialHal was gifted an aimbot to hit enemies more easily. Respawn says it would share more information soon, but as of time of writing, the studio hasn’t elaborated.

Continue reading.

The Mevo Core has improved built-in mics and works with any MFT lens.

Logitech is expanding its Mevo lineup of livestreaming cameras. The company’s new Mevo Core shoots in 4K, a big upgrade from the 1080p Mevo Start camera kit I tested a few years back. However, the trade-off is pricing as the new model will set you back three times as much for a three-camera setup. $999. So yes, this is probably for the pro streamers.

To emphasize that, the Core ships as a body only, but Logitech will sell lens bundle kits through Amazon and B&H Photo Video. You will need to buy an additional lens just to make it work. And it’s only compatible with — so there’s a high chance you’ll have to buy one.

Continue reading.

It’s like Google search on Safari all over again. Plus 15 years.

Apple is reportedly in talks with Google to integrate its Gemini AI in iPhones, according to Bloomberg. Gemini could be the cloud-based generative AI engine for Siri and other iPhone apps, while Apple’s models could be woven into the upcoming iOS 18 for on-device AI tasks.

There are regulatory concerns to consider—the Department of Justice has already sued Google over its search dominance, including the way it pays Apple and other companies to use its search engine. But given how Microsoft and OpenAI’s partnership turned the Bing search engine into something people were actually talking about, the team-up might be worth the risk.

Continue reading.

[ad_2]

Source Article Link

Tags Blackwell, chips, GPUs, NVIDIA, Powerful, Worlds

Featured

‘Algae recycled into energy’: How one of Europe’s largest data center firms wants to harness heat from GPUs and others to grow sustainable marine flora — but will it be worth it?

[ad_1]

Data centers produce a lot of waste heat that could one day be recycled and used to heat millions of homes.

Now, French data center company Data4 has partnered with the University of Paris-Saclay to launch a project that aims to use data center heat to grow algae, which can then be recycled into energy. The pilot project, set to commence early in 2024, will be trialed in the Paris region.

This initiative, led by a diverse team of experts from various fields, is driven by the French administration “Conseil Départemental de l’Essonne” and the Foundation Université Paris-Saclay. The project comes as a response to the escalating environmental impact of data centers, which have seen a 35% annual increase in data storage worldwide.

A more efficient alternative

Turning algae into heat

(Image credit: Data4)

The algae grown from the captured CO2 will be recycled into biomass to create new circular energy sources and will also be used in the production of bioproducts for other industries.

According to a feasibility study conducted with start-up Blue Planet Ecosystems, the carbon capture efficiency of this method can be 20 times greater than that of a tree.

Data4 says using the data center waste heat for the growth of algae is a more efficient alternative to the common practice of using it to warm nearby homes, which only utilizes 20% of the heat produced.

“This augmented biomass project meets two of the major challenges of our time: food security and the energy transition. This requires close collaboration between all the players in the Essonne region, including Data4, to develop a genuine industrial ecology project, aimed at pooling resources and reducing consumption in the region. Thanks to this partnership with the Fondation de l’Université Paris Saclay, we have the opportunity to draw on one of the world’s most prestigious scientific communities to work towards a common goal of a circular energy economy,” says Linda Lescuyer, Innovation Manager, Data4.

Ridiculously powerful PC with six Nvidia RTX 4090 GPUs and liquid cooling finally gets tested — there’s no game benchmarks, but plenty of tests for scientists and pros

[ad_1]

Comino, known for its liquid-cooled servers, has finally released its new flagship for testing.

The Comino Grando Server has been designed to meet a broad spectrum of high-performance computing needs, ranging from data analytics to gaming.

In a comprehensive test by StorageReview, the Grando Server, alongside a Grando Workstation variation, was put through a series of rigorous benchmarks including Blender 4.0, Luxmark, OctaneBench, Blackmagic RAW Speed Test, 7-zip Compression, and Y-Cruncher.

The server, equipped with six Nvidia RTX 4090s, AMD‘s Threadripper PRO 5995WX CPU, 512GB DDR5 DRAM, a 2TB NVMe SSD, and four 1600W PSUs, delivered impressive results, as you’d expect from those specifications.

Grando Server features advanced liquid cooling, allowing it to accommodate those six Nvidia 4090 GPUs side by side, a setup that Lyle Smith from StorageReview noted would be impossible with air cooling.

GPU-driven performance systems done right

The main difference between the Grando Server and Grando Workstation lies in their GPU and CPU capabilities. With its higher core count, the Server model is well-suited for parallel processing tasks. While still powerful, the Workstation offers fewer cores and is better suited for balanced performance across various applications.

The Grando Workstation that Smith tested featured four Nvidia A100 professional GPUs, a Threadripper Pro 3975WX processor, 512GB DDR5 DRAM, and a 2TB NVMe SSD.

In his review, Smith found the Comino Grando Server and Workstation to be “prime examples of GPU-driven performance systems done right.”

He said, “Performance-wise, the benchmark results paint a clear picture: the Grando Server excels in GPU-intensive tests like OctaneBench and Blender 4.0, highlighting its capability to breeze through high-end rendering jobs. With its tailored CPU-GPU balance, the Workstation version offers versatility for various professional applications. The speed and efficiency of both systems in managing large datasets, as evidenced by the 7-zip Compression and Y-Cruncher benchmarks, underscore their capacity for handling data-intensive operations, a critical advantage in today’s data-driven landscape.”

StorageReview approved of the Comino Grando systems so much that it rated them one of its “Best of 2024” award winners.

AMD CEO offers to help AI Startup that uses Radeon 7900 XTX GPUs — but will it backfire for Lisa Su as she seemingly supports use of consumer parts in enterprise use cases

[ad_1]

We’ve written previously about tinybox. The $15,000 AI server system is powered by AMD Radeon RX 7900 XTX graphics cards and can reportedly deliver 37% of Nvidia H100 compute performance.

It seems however, that the creators of tinybox have run into problems with bugs affecting the Radeon-based platform. After parent company tiny corp posted several tweets expressing frustration with AMD’s AI acceleration toolkit – in which it cheekily tagged AMD rivals Nvidia and Intel – AMD’s CEO, Lisa Su, stepped in, saying her team was working to fix the issues.

Unfortunately, the fixes weren’t good enough for tiny corp, which fired off another round of frustrated tweets, asking AMD to “fix their basic s*t” and suggesting the tech giant open source its firmware so that the tiny startup could do what AMD seemed incapable of – namely “fix their LLVM spilling bug and write a fuzzer for HSA”.

If AMD open sources their firmware, I’ll fix their LLVM spilling bug and write a fuzzer for HSA. Otherwise, it’s not worth putting tons of effort into fixing bugs on a platform you don’t own. https://t.co/c4I2So27YGMarch 5, 2024

The dilemma

After things got even more heated, Su tweeted “Thanks for the collaboration and feedback. We are all in to get you a good solution. Team is on it.”

While that could be good news for tiny corp – time will tell – Su could well face a backlash for essentially stepping in to support the use of its consumer products in a server aimed at the enterprise.

As jlake3 commented over on Tom’s Hardware, “tinybox is buying consumer cards instead of datacenter models, but seems to expect a datacenter SLA? They got revised firmware within 6 hours of the earliest linked tweet and a call with engineering the next day, which is more than I’d think a startup buying a bunch of gaming cards would qualify for when there are actual paying enterprise clients, and they appear to be having a public meltdown that AMD isn’t doing more for a startup with less than 100 servers built and none yet shipped (and expressly avoiding using pro GPUs).”

He also made another good point. “As for Nvidia, their EULA forbids using GeForce products for datacenter CUDA applications, so they’d definitely not be dedicating anyone to talk to tinybox in this situation, except maybe a lawyer.”

Firm headed by legendary chip architect behind AMD Zen finally releases first hardware — days after being selected to build the future of AI in Japan, Tenstorrent unveils Grayskull, its RISC-V answer to GPUs

[ad_1]

Tenstorrent, the firm led by legendary chip architect Jim Keller, the mastermind behind AMD‘s Zen architecture and Tesla’s original self-driving chip, has launched its first hardware. Grayskull is a RISC-V alternative to GPUs that is designed to be easier to program and scale, and reportedly excels at handling run-time sparsity and conditional computation.

Off the back of this, Tenstorrent has also unveiled its Grayskull-powered DevKits – the standard Grayskull e75 and the more powerful Grayskull e150. Both are inference-only hardware designed for AI development, and come with TT-Buda and TT-Metalium software. The former is for running models right away, while the latter is for users who want to customize their models or write new ones.

The Santa Clara-based tech firm’s milestone launch comes hot on the heels of a partnership with Japan’s Leading-edge Semiconductor Technology Center (LSTC). Tenstorrent’s RISC-V and Chiplet IP will be used to build a state-of-the-art 2nm AI Accelerator, with the ultimate goal of revolutionizing AI performance in Japan.

By the power of Grayskull!

The Grayskull e75 model is a low-profile, half-length PCIe Gen 4 board with a single Grayskull processor, operating at 75W. The more advanced e150 model is a standard height, 3/4 length PCIe Gen 4 board containing one Grayskull processor operating at up to 200W, and balancing power and throughput.

Tenstorrent processors comprise a grid of cores known as Tensix Cores and come with network communication hardware so they can talk with one another directly over networks, instead of through DRAM.

The Grayskull DevKits support a wide range of models, including BERT for natural language processing tasks, ResNet for image recognition, Whisper for speech recognition and translation, YOLOv5 for real-time object detection, and U-Net for image segmentation.

The Grayskull e75 and e150 DevKits are available for purchase now at $599 and $799, respectively.