rolvram
1 día hace
Dell unveils new AI servers powered by Nvidia chips to boost enterprise adoption
12:30:00 PM ET, 05/19/2025 - Reuters
May 19 (Reuters) - Dell Technologies on Monday unveiled new servers powered by Nvidia's Blackwell Ultra chips, aiming to capitalize on the booming demand for artificial intelligence systems.
The servers, available in both air-cooled and liquid-cooled variations, support up to 192 Nvidia Blackwell Ultra chips but can be customized to include as many as 256 chips.
These servers can train AI models up to four times faster than previous models, Dell said.
The pricing of these products will be "competitive," Arthur Lewis, President of Dell's Infrastructure Solutions Group, told Reuters, adding that "there's a lot of interest on what's next."
Dell and Super Micro Computer have benefited from the growing demand for servers designed to handle the computer-heavy AI tasks, but the high cost of producing the systems and tough competition have pressured their margins.
Dell in February forecast a decline in adjusted gross margin rate for fiscal 2026, while Super Micro projected fourth-quarter revenue below estimates earlier this month as tariff-driven economic uncertainty pressures its performance.
Dell will focus on increasing sales of networking and storage products to ensure the "right level of profitability," Lewis said.
The company's new servers will also support Nvidia's upcoming Vera central processing units, which will succeed the chip designer's Grace server processor.
The AI server maker plans to support Nvidia's Vera Rubin chips, set to follow the Blackwell series.
Dell also introduced a 'Pro Max Plus' laptop designed for AI development on Monday, featuring a neural processing unit that allows engineers to process large AI models directly on the device without relying on cloud services. (Reporting by Jaspreet Singh in Bengaluru; Editing by Shreya Biswas)
rolvram
1 día hace
US Chip, Software Giants Enter 'Watershed Moment' in AI Shift to Mideast, Wedbush Says
11:31:21 AM ET, 05/19/2025 - MT Newswires
11:31 AM EDT, 05/19/2025 (MT Newswires) -- Nvidia (NVDA), Microsoft (MSFT) and other US semiconductor and software giants face a "watershed moment" as the artificial intelligence revolution expands to the Middle East, Wedbush Securities said Monday in a note.
"The market opportunity in Saudi Arabia and UAE alone could over time add another $1 trillion to the broader global AI market in the coming years, and this dynamic is not being priced into the market and tech names," Wedbush said.
President Donald Trump's Middle East trip culminated in deals for the delivery of 18,000 Nvidia chips to Saudi Arabia for a 500-megawatt data center and for the largest AI data center outside the US in Abu Dhabi, the report said.
The announcements mark the "start of a new era of growth for the US tech sector and a gamechanger for the industry," said the investment firm.
The "massive AI buildout" in Saudi Arabia and the UAE might open up a "huge opportunity" and total addressable market in the next decade for Nvidia, Palantir Technologies (PLTR), Microsoft, Amazon (AMZN), Alphabet (GOOG,GOOGL), Tesla (TSLA) and other companies, Wedbush said.
The tech boom in the Middle East "also speaks to this global game of high stakes poker being played by the Trump administration" as China faces US chip restrictions, the report said. " China is clearly taking notice and this adds to the drama" of broader US-China trade talks, Wedbush said.
rolvram
2 días hace
https://seekingalpha.com/author/danil-sereda#source=url_first_level%3Aarticle%7Csection%3Aauthor_brief%3Adanil-sereda%7Csection_asset%3Aauthor_brief%7Cauthor_namehttps://seekingalpha.com/checkout/mp_1415#source=url_first_level%3Aarticle%7Csection%3Aauthor_brief%3Adanil-sereda%7Csection_asset%3Aauthor_brief%7Cinvesting_group_badge(13min)https://seekingalpha.com/author/danil-sereda/analysis?ticker=NVDA#hasComeFromMpArticle=true#source=section%3Amain_content%7Cbutton%3Abody_link of https://seekingalpha.com/symbol/NVDA#hasComeFromMpArticle=true#source=section%3Amain_content%7Cbutton%3Abody_linkspans a couple of years here on Seeking Alpha. The last time I covered the stock was https://seekingalpha.com/article/4768709-nvidia-rare-buying-opportunity-following-gtc-event#hasComeFromMpArticle=true#source=section%3Amain_content%7Cbutton%3Abody_link, when I reiterated my "Buy" rating with a thesis that the firm's attempts to enter the promising AI robotics market should massively expand its TAM and loosen semiconductor-specific cyclicality, and that the GTC event gave me a fresh confirmation that Nvidia is very "likely to step aside from its main semi business" eventually and open up a new revenue stream in the next few years. Below you can see my coverage history snapshot:Today, I'd like to preview Nvidia's upcoming fiscal Q1 FY2026 earnings release. From the data I have on hand, I strongly believe we'll see another strong print, beating the currently somewhat depressed earnings estimates. What's more important than just strong actual figures for the Q1 will likely be optimistic guidance from the management team, with hopefully some updates on the robotics business. Despite some risks surrounding the biggest semiconductor company in the world, I think its moat is still there, so I reiterate my "Buy" a few days ahead of its earnings release day (https://seekingalpha.com/symbol/NVDA/earnings#hasComeFromMpArticle=true#source=section%3Amain_content%7Cbutton%3Abody_link).My previous take on NVDA already included its Q4 FY2025 review, so for this article, let me just briefly touch upon what was really important at the time. The company showed https://seekingalpha.com/symbol/NVDA/earnings/eps-surprise-summary?period=quarterly#hasComeFromMpArticle=true#source=section%3Amain_content%7Cbutton%3Abody_link with sales at $39.3 billion (+78% YoY) and earnings per share at ~$0.89, which was an increase of 71.1% YoY - both figures easily beat the consensus estimates, although with a weakening momentum compared to the last few quarters.What caught my eye immediately was Nvidia's data center revenues of ~$35.6 billion (+93% YoY and +16% QoQ), which saw such a massive move up from a quicker-than-expected ramp of its Blackwell architecture, which yielded ~$11 billion in revenue in a single quarter; plus we saw a continued strength in Hopper architecture sales. It's clear that multiple scaling laws are now driving AI compute demand: pre-training scaling for multimodality and reasoning-enhanced data, post-training scaling using reinforcement learning (which will likely demand orders of magnitude more compute than pre-training, in my understanding), and "test time compute or reasoning, long thinking, inference scaling", as the management put ithttps://seekingalpha.com/article/4762511-nvidia-corporation-nvda-q4-2025-earnings-call-transcript#hasComeFromMpArticle=true#source=section%3Amain_content%7Cbutton%3Abody_link. As Huang noted further on the call, "long thinking reasoning AI can require 100 times more compute per task compared to one-shot inferences", and if so, the demand side is likely to stay as strong as it is today (at least). And the Blackwell architecture should be the main driving force to meet that potential demand because it was mainly architectures for reasoning AI inference (Blackwell can handle ~25x higher token throughput at a 20x lower cost vs. Hopper 100).According to the updated research study by MarketsAndMarkets, the global https://www.marketsandmarkets.com/Market-Reports/data-center-gpu-market-18997435.html?utm_source=prnewswire.com&utm_medium=referral&utm_campaign=datacentergpumarket is projected to reach ~$120 billion this year and ~$228.04 billion by 2030, so the resulting CAGR will amount to 13.7% if these calculations are close to reality. At the same time, Goldman Sachs analysts forecast that the balance of demand and supply should tighten further to 2026, only then moderating slightly as more DCs come online:So I see no reason to expect any negative comments about the demand side from Jensen Huang and his team on the upcoming call.The situation with China and U.S. export controls adds some uncertainties, of course, with Nvidia's data center sales in China "remaining well below levels seen at the onset of export controls", according to the CFO's comments. But Nvidia seems to be navigating the existing restrictions quite well. I think they understand the significance of potential Chinese demand as both nations fight for AI dominance, and I view their plans of developing China-specific chips (such as the downgraded H20 AI chip to comply with U.S. rules) as a smart move to fulfill that regional demand, too. https://www.ft.com/content/c886a4c0-da75-4ea7-8230-6ffd18815fa4 that Nvidia is going to establish a new R&D center in Shanghai, and while it might seem like a risky move amidst the trade war, I believe it's actually a smart move (if they keep complying with the U.S. government, and that's the likely scenario).So again, I'd expect some clarity on their China plans, and if they align with my expectations, the stock price should theoretically react positively after such comments.Scrolling through banking research papers, I got one from UBS (proprietary source), which is a preview of NVDA's upcoming Q1 results. The bank's analysts wrote that they expect the H20 ban charge (Nvidia took a $5.5 billion charge for its upcoming quarterly results, https://www.reuters.com/technology/nvidia-expects-up-55-billion-charge-first-quarter-2025-04-15/#:~:text=April%2015%20%28Reuters%29%20%2D%20Nvidia,of%20its%20most%20popular%20chips.) might impact its Q1 non-GAAP gross margin. That's the part I'd be watching particularly closely as it's seemingly becoming a major risk factor (after the Q4 print, when the GP margin kept sliding lower on a QoQ basis, the stock price reacted with a sharp mode down eventually):But even if we see some continuation in GP margin contraction in Q1, I think that the management's commentary to expect that $5.5 billion charge helped NVDA price in some negativity already. Plus, Nvidia's pricing power and value proposition seem to remain strong as their new, more advanced chips should bring in higher ASP and keep the GP margins above their long-term norms. Anyway, the management itself expects margins to be at 71% (+/- 50 bps) in Q1, gradually improving to the mid-70% range as "production scales and cost efficiencies materialize".Additionally, Nvidia guided for ~$43 billion in revenues for Q1 (+/- 2%), and it's aligning well with the current consensus estimates, which have been revised lower since NVDA's fiscal Q4 FY2025 release.What I find interesting is that the GP margin concerns flew through most analysts' income statement forecast models and resulted in some massive downward revisions for the next few quarters in a row:Coming back to UBS's take on the upcoming Q1 print, I see that they expect the top-line growth to reaccelerate in 2H FY2026 "as GB300 racks begin shipping and NVIDIA potentially resumes shipments of a modified Blackwell-based SKU to China", but at the same time, they do see a lower GP margin at 58.5% in Q1 (non-GAAP) by their calculation if H20 charges flow through non-GAAP (recall, the company guided for 71%). I think they're too conservative on this assumption, but even with it in mind, they wrote that Nvidia looks quite tempting on the valuation front (they set a price target at ~$175/share:Another major investment bank, whose reports I tend to go through quite often - Morgan Stanley - agreed on NVDA's undervaluation. They see NVDA's fair price target at $160/share, selecting the stock as their "Top Pick" in semis:I myself think that Nvidia is quite tempting at its current price levels. If we take a look at the priced-in P/E multiple contraction for the next year, we'll see that the stock is currently trading close to the lowest point over the past 10 years, based on YCharts data:I know they're operating in a cyclical industry, so it's completely OK to expect a sharp valuation multiples contraction on the horizon. But I'm not sure that a) the AI/DC cycle is turning just yet, and b) Nvidia's revenue structure will remain as dependent on just designing/selling GPUs as it is to date (again, robotics should eventually keep the valuation premium higher for longer, in my view).So with all that in mind, I suggest taking both UBS's and Morgan Stanley's price targets into account, deriving a midpoint as a more realistic base-case scenario. It arrives at $167.5/share, giving me an upside potential of 23.7% to the current price.I think that the main risk to my bullish reasoning and "Buy" reiteration right before the Q1 print lies in my assumptions regarding Q1 GP margins. The charge of $5.5 billion, which was admitted by the management team, leaves a huge risk to the bottom line. If Nvidia misses this time, the market's reaction may be severely negative as some of the current valuation hangs on hopes of sustainably higher margins compared to the company's history.Also, if the stock's P/E keeps naturally sliding down because of cyclicality and lack of clarity on the robotics side of Nvidia's business, then at 20x FY2027 earnings and the current EPS consensus, the stock would be ~16.5% overvalued. Let that be the bear-case scenario if they miss.Despite the above risks, I think that Nvidia's fundamental growth drivers are currently more powerful than ever before. If the AI revolution is still in its early innings, as I expect it, then the cyclical turn is well beyond calendar 2025-2026. In this case, it's logical to expect a gradual increase in Nvidia's stock price as it's going to reprice the ongoing tailwinds.After the NVDA stock fell by over 45% since its January 2025 ATH point, it has managed to almost fully recover; I think this could be just the beginning of an eventual rip even higher.Based on all that, I reaffirm my "Buy" rating on NVDA a few days before its fiscal Q1 FY2026 earnings release.
doc2016
3 días hace
platform implementation strategies
ChatGPT said:
Implementing a robust platform that integrates CPaaS (Communications Platform as a Service) and UCaaS (Unified Communications as a Service) to facilitate collaboration between customer-facing agents, internal subject matter experts (SMEs), and generative AI systems requires a strategic approach. Here are key strategies to consider:
1. Phased Rollout Deployment
Implementing the platform in stages allows for manageable adoption and minimizes disruptions. Begin with a pilot program involving a specific department or team, gather feedback, and make necessary adjustments before a company-wide rollout. This approach ensures smoother transitions and better user acceptance.
Klaxoon
2. Parallel Adoption Strategy
Running the new platform alongside existing systems for a period allows for validation and comparison. This strategy helps in identifying issues early and provides a safety net, ensuring business continuity during the transition.
Klaxoon
3. Vendor-Supported Deployment
Leveraging the expertise of platform vendors can streamline the implementation process. Vendors can provide training, best practices, and support, ensuring that the platform is configured optimally to meet organizational needs.
4. Establish Clear Governance
Defining clear accountability structures and implementing robust monitoring systems are crucial. Establishing comprehensive security measures protects digital assets and ensures compliance with regulatory standards.
Accelare
5. Embrace Organizational Change
Implementing a new platform often requires significant changes in processes and culture. Fostering a culture of innovation, agility, and continuous learning ensures that all stakeholders are aligned and committed to the transformation.
Upcore
6. Invest in Talent and Skills Development
Building and maintaining advanced digital platforms necessitates a diverse set of skills. Investing in attracting and retaining top talent, and providing ongoing training, ensures that the workforce is equipped to leverage the new platform effectively.
Upcore
7. Leverage Partnerships and Ecosystems
Recognize that not all capabilities need to be developed in-house. Identifying strategic partners that can complement your capabilities and provide specialized expertise or services can enhance the platform's effectiveness.
Upcore
8. Prioritize Data and Analytics
Implementing robust data management strategies and investing in advanced analytics tools enables data-driven decision-making. This approach ensures that the platform evolves based on continuous feedback and changing preferences.
Upcore
Accelare
9. Embrace Agile and DevOps Practices
Adopting agile methodologies and DevOps principles accelerates the development and deployment of new features and updates. Encouraging cross-functional collaboration and automation streamlines processes and improves efficiency.
Upcore
10. Continuously Iterate and Improve
Digital platforms are dynamic and require constant refinement. Gathering feedback from users, analyzing performance data, and iterating on digital offerings ensure that the platform remains relevant and competitive.
Upcore
By following these strategies, organizations can effectively implement a platform that enhances collaboration between customer-facing agents, internal SMEs, and generative AI systems, ensuring a resilient and adaptive communication infrastructure.
If you need further assistance in tailoring these strategies to your specific organizational context or selecting appropriate platforms, feel free to ask!
Monitoree múltiples cotizaciones en tiempo real de los principales intercambios, como la Bolsa de Valores de Londres, NASDAQ, NYSE, AMEX, Bovespa y más.
LoveThemStocks
6 horas hace
layton
22 horas hace
StarryChimera74
2 días hace
MillionaireTradersAlliance
2 días hace
Nebuchadnezzar
3 días hace
Jetmek_03052
3 días hace
JJ8
4 días hace
Hot Features
Premium
