Nvidia Smashes $81.6B Record, Dell PowerRack Goes Deep, HPE Q2 Earnings June 1

By Network-Switch.com  |  May 26, 2026

Table of contents
Technician Managing Optical Fiber and Coaxial Cables in a Dell Data Center Cabinet

1. Nvidia Q1 FY2027 Results (May 20): $81.6B Revenue, $14.8B Networking, Q2 Guided at $91B

Nvidia reported record Q1 FY2027 results on May 20 that beat Wall Street consensus on every key metric. Total revenue reached $81.6 billion - up 85% year-over-year and 20% sequentially - against analyst consensus of approximately $78 billion. Data Center revenue came in at a record $75.2 billion, up 92% year-over-year and 21% sequentially, driven by the ramp of Blackwell 300 products and demand for InfiniBand, Spectrum-X Ethernet, and NVLink solutions. Non-GAAP EPS of $1.87 beat the $1.77 consensus by 5.6%.

The networking line deserves separate attention: Data Center networking revenue hit a record $14.8 billion - up 199% year-over-year and 35% sequentially, reflecting the continued ramp of NVLink compute fabric for GB200 and GB300 systems alongside growth across InfiniBand and Ethernet platforms. For the first time, hyperscale customers (Microsoft, Meta, Google, Amazon, Oracle) represented approximately 50% of Data Center revenue - with the other 50% coming from AI Clouds, industrial, enterprise, and sovereign customers. CEO Jensen Huang described this diversification as the "agentic AI inflection point," adding: "The buildout of AI factories - the largest infrastructure expansion in human history - is accelerating at extraordinary speed." Q2 FY2027 guidance was set at $91 billion, plus or minus 2%, implying 93% year-over-year growth. No China Data Center compute revenue is included in the Q2 outlook. Nvidia also raised its quarterly dividend 25x, from $0.01 to $0.25 per share, and authorized an additional $80 billion share repurchase program. Record free cash flow of $49 billion was generated in the quarter.

2. Dell Technologies World Wrap-Up: PowerRack Networking Delivers 800+ Tb/s, 5,000 AI Factory Customers

Dell Technologies World 2026, which ran May 18-21 in Las Vegas, generated a sustained stream of AI infrastructure announcements across all four days. The centerpiece - Dell PowerRack - was detailed further in technical sessions after Day 1's keynote reveal. The PowerRack networking configuration employs eight Dell PowerSwitch SN6600-LD switches - a branded version of Nvidia's Spectrum SN6600-LD Ethernet switch - delivering over 800 Tb/s of east-west switching capacity per rack. The system is pre-validated at the factory and managed via the Dell Integrated Rack Controller, shipping in both 19-inch and 21-inch rack form factors with a choice of air or liquid cooling.

Beyond hardware, Dell disclosed that it added 1,000 new customers to its AI Factory with NVIDIA in a single quarter, bringing the total to 5,000 enterprise deployments. Reference customers named publicly include Eli Lilly, Honeywell, Ascension, and Samsung - organizations running AI workloads on infrastructure they own and control rather than in the cloud. Michael Dell framed the strategic shift on stage: "In companies across every industry, AI is accelerating from proof-of-concept into production. It's flipping the traditional buy-vs-build equation." Dell also introduced the Dell AI Ecosystem Program, a new structured framework for integrating AI software partners - including Google, Hugging Face, OpenAI, Palantir, ServiceNow, Reflection, and SpaceXAI - into validated, enterprise-ready configurations on top of Dell AI Factory infrastructure.

3. Dell PowerStore Elite: 5.8 Effective PB in 3U, Triples Performance Over Prior Generation

Announced at Dell Technologies World 2026, the Dell PowerStore Elite is Dell's new flagship enterprise storage platform, delivering a triple improvement in both performance and density compared to the previous PowerStore generation. The system condenses up to 5.8 effective petabytes of capacity into a single 3U device - a density figure Dell states is industry-leading based on publicly available third-party documentation as of March 2026. It supports SSDs as large as 245 TB each, and the underlying Exascale Storage hardware platform is designed to support file, object, and parallel-file engines on one reusable hardware base - a 3-in-1 storage architecture. Block storage support is targeted for availability in the first half of calendar year 2027.

The PowerStore Elite is a direct response to the data demands of AI inference at enterprise scale: as organizations move AI workloads from training (compute-heavy, relatively low storage I/O) to inference (continuous, low-latency, high-throughput data retrieval), storage architecture has become a performance-critical variable. Dell's internal benchmarks cite 150 GB/sec read performance for its Lightning File System on the same underlying hardware - a figure the company compares against 20x lower-performing traditional flash scale-out systems. PowerStore Elite is available now, while the Dell PowerRack for Exascale Storage rack-scale configuration is targeted for availability in the second half of 2026.

4. Dell Deskside Agentic AI: Trillion-Parameter Models at the Edge Without Cloud

One of the most strategically significant announcements at Dell Technologies World 2026 was not a data center product: Dell introduced Dell Deskside Agentic AI, a workstation-class edge inference platform capable of running trillion-parameter AI models locally - without sending data to external cloud environments. The announcement reflects the convergence of two previously separate markets: enterprise AI infrastructure (historically a data center conversation) and the commercial workstation (historically a client computing conversation).

Dell Deskside Agentic AI is built on Nvidia RTX Pro Blackwell Server Edition GPUs and is specifically designed for enterprise use cases requiring data sovereignty, low latency, and air-gapped AI operation - healthcare, defense, legal, and financial services being the primary targets. The product positions Dell's "One Dell Way" organizational unification (announced May 3) in concrete product terms: the same Dell infrastructure platform now spans from the edge workstation to the hyperscale AI factory rack. CEO Michael Dell has consistently argued that "AI's future follows data," meaning enterprises will need AI infrastructure wherever their data lives - and this product is the clearest expression of that thesis in hardware.

5. HPE Q2 FY2026 Earnings Webcast Set for June 1 - Networking Revenue Expected to Grow 142-152% YoY

Hewlett Packard Enterprise confirmed on May 19 that it will host a live audio webcast on June 1, 2026 to discuss Q2 FY2026 financial results for the period ending April 30. HPE stock closed at $33.80 on May 20 - up 3.6% on the day Nvidia reported, reflecting positive AI infrastructure sentiment spillover - and has now rallied significantly year-to-date as investors reassess the company's networking and AI infrastructure positioning.

HPE guided Q2 FY2026 networking revenue growth of 142% to 152% year-over-year on a reported basis when it reported Q1 results in March - driven by the full quarter consolidation of the Juniper Networks acquisition and continued AI data center networking demand. The company entered Q2 with a record AI systems backlog of $5 billion, composed primarily of enterprise and sovereign orders. Management noted on the Q1 call that Q3 is expected to be its largest AI revenue quarter for the full fiscal year, with profitability weighted toward Q4 per historical linearity. CFO Marie Myers guided for non-GAAP EPS of $0.51 to $0.55 in Q2, and the company raised full-year non-GAAP EPS guidance to $2.30-$2.50, representing 32-40% operating profit growth. Free cash flow guidance was set at a minimum of $2 billion for fiscal 2026 - a significant swing from negative $877 million in Q1 FY2025.

6. Arista Networks at JP Morgan TMC Conference (May 19): AI and Campus Networking Outlook Reaffirmed

On May 19, Arista Networks' Chief Customer Officer Ashwin Kohli and Senior VP of Cloud and AI Networking Tyson Lamoreaux represented the company at the JP Morgan 2026 Global Technology, Media and Communications Conference in a 35-minute session. The appearance came two weeks after Arista's strong Q1 2026 results ($2.709 billion revenue, +35.1% YoY, AI target raised to $3.5 billion for 2026) and served as an opportunity to reinforce the company's full-year outlook ahead of the Needham Technology Conference on May 14 and William Blair's Growth Stock Conference later in May.

Key themes from Arista's investor conference circuit this month: the continued ramp of 800G deployments at Microsoft and Meta; the maturation of the campus networking segment following the VeloCloud SD-WAN integration, which Arista has guided for 60% growth in 2026; and the competitive positioning of open Ethernet standards versus Nvidia Spectrum-X in hyperscale AI back-end switching. Analysts following the sessions noted that Arista's management reiterated confidence in the $3.5 billion AI networking revenue target for 2026 and maintained that Ethernet's platform-agnostic advantage - the ability to serve both Nvidia and AMD-based GPU clusters with the same switch fabric - remains a structural differentiator in a market where multi-vendor AI deployments are growing.

Editor's Summary

The week of May 19-25, 2026 will be remembered primarily for Nvidia's Q1 FY2027 print - the largest quarterly revenue report in semiconductor history at $81.6 billion, with data center networking alone reaching $14.8 billion (+199% YoY). The results confirm that AI infrastructure spending is not decelerating: hyperscale customers maintained their pace while sovereign and enterprise deployments now represent half of Nvidia's Data Center revenue for the first time, a structural broadening of the demand base that reduces reliance on any single customer category. Dell's DTW wrap-up showed that PowerRack - with 800+ Tb/s Nvidia Spectrum-based switching per rack - is Dell's definitive answer to the question of how enterprises deploy full-scale AI infrastructure on-premises. HPE's June 1 earnings call, approaching with a $5 billion AI backlog and 142-152% guided networking growth, will be the next major data point in the AI infrastructure earnings cycle. And Arista's investor conference appearances this week signal a company in execution mode, not pivot mode - the $3.5 billion AI networking target for 2026 remains intact.

Network-Switch.com is a global professional distributor of networking equipment from Cisco, Huawei, Ruijie, H3C, and our own NS brand - including switches, routers, firewalls, wireless APs, optical modules, and fiber patch cables. Our CCIE, HCIE, H3CIE, and RCNP certified engineers deliver end-to-end technical support and complete network solutions for enterprise, campus, and large-scale infrastructure deployments worldwide.

Visit us: https://network-switch.com

Frequently asked questions (FAQs)

What was Nvidia's Q1 FY2027 revenue and data center networking revenue?

Nvidia reported record Q1 FY2027 revenue of $81.6 billion (up 85% year-over-year). Data Center networking revenue - covering InfiniBand, Spectrum-X Ethernet, and NVLink solutions - reached a record $14.8 billion, up 199% year-over-year and 35% sequentially. Q2 FY2027 guidance was set at $91 billion.

What is Dell PowerRack and how much switching capacity does it provide?

Dell PowerRack is a turnkey, rack-scale AI infrastructure system announced at Dell Technologies World 2026, integrating compute, networking, storage, and cooling in a single pre-validated unit. The PowerRack networking configuration uses eight Dell PowerSwitch SN6600-LD switches (based on Nvidia Spectrum silicon) and delivers over 800 Tb/s of east-west switching capacity per rack. It can go from delivery to running live AI or HPC workloads in as little as 6.5 hours. Networking availability is September 2026.

What is the Dell PowerStore Elite and when is it available?

Dell PowerStore Elite is Dell's new flagship enterprise storage platform, delivering 5.8 effective petabytes of storage in a 3U device - triple the density and performance of the previous PowerStore generation. It supports SSDs up to 245 TB each and will support file, object, and parallel-file storage on a single hardware platform. It is available now, with block storage support targeted for the first half of calendar year 2027.

When does HPE report Q2 FY2026 earnings and what is expected?

HPE will report Q2 FY2026 financial results via a live webcast on June 1, 2026, covering the quarter ended April 30, 2026. HPE guided networking segment revenue growth of 142% to 152% year-over-year for Q2, driven by the full consolidation of Juniper Networks and strong AI data center demand. The company entered Q2 with a record AI systems backlog of $5 billion.

How is Arista Networks positioned in the AI networking market in 2026?

Arista Networks holds the number-one market share position in high-speed data center Ethernet switching above 10 GbE. The company reported Q1 2026 revenue of $2.709 billion (+35.1% YoY) and raised its full-year 2026 AI networking revenue target to $3.5 billion. Arista's key differentiator is platform-agnostic Ethernet switching that supports both Nvidia and AMD GPU clusters with the same infrastructure - a growing advantage as enterprise AI deployments increasingly use multiple GPU vendors.

Sources