$NVDA

Nvidia touts AI data centre software to boost output

Nvidia presented results for its DSX software platform, which optimizes AI data center power management. DSX Flex shifts workloads during grid strain, while DSX MaxLPS reallocates power to boost output. Tests showed a 24% increase in throughput and 23% improvement in performance per watt. Nvidia aims to address power constraints in AI data centers with DSX's holistic approach, including simulation tools and reference designs.

Original reporting
Published Sep 18, 2026, 4:45 PM UTC
Analysis
AlphAI AI DeskAI-generated
Added to AlphAI Sep 18, 2026, 5:30 PM UTC. Informational, not investment advice.
How this was made
AlphAI summarizes source reporting and applies a structured AI analysis for relevance, timing, sentiment and ticker impact. Always verify material claims with the original publisher.
Nvidia touts AI data centre software to boost output — source image
Decision brief

The 30-second read

$NVDABullishMed
01

Why it matters

The disclosed performance improvements could enhance Nvidia's competitive edge in AI infrastructure, potentially supporting higher sales of its HGX GPUs and related services.

02

Market read

First public data on Nvidia's DSX efficiency claims may influence investor sentiment on Nvidia and related AI infrastructure stocks.

03

What to watch

Adoption depends on utility participation and customer willingness to integrate DSX software.

Relevance 7/10Novelty 7/10Timing: today

Background

Nvidia introduced DSX, a suite of software tools aimed at improving power utilization in AI data centres, with early results from a Santa Clara facility and Lambda cloud provider.

Company-level read

Ticker impact

$NVDABullishMedium confidence
Context

Nvidia disclosed early deployment results for its DSX AI data centre software, showing 24% token throughput increase and 23% performance‑per‑watt improvement.

Expected impact

Potential modest upside as investors price in new software capability.

Evidence & confidence

First‑time disclosure of measurable efficiency gains; market may view as incremental but supportive of Nvidia's AI leadership.

Market effects

Highlights growing demand for power‑efficient AI data‑centre solutions, benefiting hardware and software providers.

Shows Silicon Valley Power's demand‑response program can be a model for US utilities.

Positions Nvidia as a leader in AI‑centric power‑management globally.

Counterpoint

Efficiency gains may be modest relative to Nvidia's valuation; investors may already price in AI growth.

Key entities

  • Nvidia

    Developer of DSX AI data centre software and HGX B200 GPU servers.

  • Lambda

    Cloud provider that validated DSX performance gains.

  • Emerald AI

    Provider of Conductor software used in the demand‑response test.

Related articles

HighAI 9/10

Before Sale to Softcat, GDT Tapped Into Nvidia-Fueled Neocloud Gold Rush

Softcat acquired GDT for $1.05 billion, gaining a U.S. systems integration business with a strong neocloud presence. GDT's Chris Kapusta attributes success to 30 years of experience and Nvidia's financial support for neoclouds. Gartner predicts neoclouds will represent 20% of the $267 billion AI cloud market by 2030. Nvidia's investments help neoclouds access AI infrastructure, facilitating GDT's deals.

$AVGOHighAI 8/10

Nvidia’s Not Going to Be King for Life, So I Buy This Now

Broadcom (AVGO) is gaining market share in AI semiconductors, reporting a 221% revenue surge to $16.7B last quarter. The company ships custom XPUs to six hyperscalers, offering cost advantages over NVIDIA (NVDA). Broadcom targets $230B in fiscal 2028 AI revenue, contrasting with NVIDIA's supply constraints and higher valuation. Broadcom's dividend policy includes 15 consecutive annual increases.

$NVDAMedAI 8/10

These Magnificent Seven Members Still Look Cheap

Microsoft (MSFT) and NVIDIA (NVDA) are highlighted for attractive valuations despite strong growth. NVIDIA reported $96.2B revenue, up 106% YoY, with Data Center revenue at $89.0B. Microsoft's cloud revenue grew 32% YoY. Both trade at discounts to their 5-year medians, with strong earnings growth expected. Apple (AAPL) is noted as a stable alternative.

$NVDAMed

Nvidia expects to double chip sales next year

Nvidia's CEO Jensen Huang expects the company to double chip sales next year due to strong AI demand. He cited AI's significant impact on industries and economies. Nvidia recently projected 70% revenue growth for the fiscal year ending January 2028, potentially reaching $673 billion in annual revenue. The company's products include data center GPUs, CPUs, and AI-related hardware. Huang also attended an AI summit to discuss safety concerns.

$NVDAMed

Nvidia CEO Sends Strong Signal to Investors

Nvidia (NVDA) CEO Jensen Huang expects to sell twice as many chips next year, citing strong AI demand. The company projects 70% growth in fiscal 2028, with revenue around $673 billion. Huang's forecast suggests sustained AI investment across industries, but execution risks remain.

$NVDAHighAI 8/10

Here’s When Nividia’s Dominance Fades, Opening Broadcom’s Era of Dominance

NVIDIA (NVDA) and Broadcom (AVGO) reported strong AI-driven earnings. NVDA's revenue rose 105.8% YoY, with data center revenue up 117%. AVGO's AI semiconductor revenue grew 221% YoY. AVGO claims its TPU v8i matches NVDA's Vera Rubin at half the cost, with $230B in AI supply commitments by 2028. AVGO trades at a lower forward P/E (18) than NVDA (23).