# data center revenue — X 热门讨论 (2026-10-01 01:10 UTC)
## @VanquishTrader (VanquishTrader) · 09-30 22:34 · ♥58 ↻4 💬4 WHY $MU COULD BE A $2,000 STOCK
My biggest takeaway from Micron’s earnings is that the path toward a $2,000 stock has much less to do with another blowout quarter and much more to do with the market eventually realizing this business may deserve a completely different earnings multiple. Revenue reached roughly $54.2 billion with adjusted EPS of $33.42 and an 87% gross margin, while Q1 guidance moved to roughly $61.5 billion of revenue and $38 of EPS. What makes that even more impressive is Q4 had an extra week, so on a normalized weekly basis Q1 implies roughly 22% sequential growth. Even more important, management said Q1 should be the gross margin floor for FY27, which tells me the business can keep growing even as the enormous pricing increases from earlier in the cycle begin to moderate.
The reason I think this cycle can look so different is AI is changing where Micron’s demand comes from and how much memory every system needs. HBM is absorbing enormous wafer capacity and supporting DRAM pricing, while conventional DRAM benefits from larger AI servers and enterprise NAND is beginning to see a completely new demand curve from KV cache and persistent context as agents run longer and store more information. Data center SSD revenue alone approached $10 billion after growing more than tenfold year over year, which is a huge signal that the AI memory story is expanding well beyond HBM. At the same time Micron now has 26 multiyear strategic agreements representing more than 35% of revenue through 2030, giving the company much better volume visibility and a potentially much higher earnings floor than investors are used to seeing from memory.
That higher floor becomes even more powerful when you combine it with the amount of cash Micron is now producing. The company generated roughly $33.2 billion of adjusted free cash flow in the quarter and management plans to increase capital returns once its CHIPS restrictions roll off, with the long term goal of returning 100% of excess cash to shareholders. What stands out to me is Micron is generating this cash without responding to the boom by massively overbuilding, because capex remains relatively controlled compared with revenue and operating cash flow. That combination of stronger margins, better customer visibility, disciplined supply and potentially enormous buybacks can create a very different per share earnings profile than previous memory peaks.
That is ultimately why I think Micron has a credible path toward $2,000 over time since the market still largely treats the company like a cyclical memory manufacturer whose earnings eventually collapse when supply catches demand, but AI is increasing memory intensity across HBM, DRAM and enterprise storage while long term agreements and disciplined capacity are making the downside of the cycle potentially much less severe. If Micron can sustain a meaningfully higher earnings floor while continuing to shrink the share count with excess cash, the rerating does not require heroic assumptions. The biggest risk remains new capacity eventually catching demand, but if FY27 proves that earnings can stay elevated even as pricing growth slows, I think investors are going to have to rethink what normalized earnings actually look like for Micron. https://x.com/VanquishTrader/status/2105425979233726767
## @athuinvests (Athu) · 09-30 20:33 · ♥35 ↻4 💬4 $MU EARNINGS IS INSANE!
Micron has no "line of sight to when supply and demand will return to balance."
RPO is ~$150B across 26 take-or-pay SCAs, which Micron calls conservative because it uses committed volumes and minimum pricing. Customers have put up $32B in commitments, mostly cash deposits.
(Numbers) - Revenue: $54.23B (+379% YoY) - EPS: $33.42 (+1,003% YoY) - Gross margin: 87.0% (46% a year ago)
(Technology) - DRAM: $39.8B (+343% YoY) - NAND: $14.1B (+526% YoY) - Core data center: $18.0B (+1,042% YoY) - Cloud memory: $16.3B (+258% YoY)
DRAM prices rose in the high-teens QoQ and NAND about 30%.
(Guide) Q1 FY27 guide: - revenue $60B-$63B, - EPS $37.15-$39.15, - gross margin ~86.25%
Q1 is the gross margin floor for FY27. It rises after that, with a more moderate pace of price increases.
The vast majority of calendar 2027 HBM supply is already agreed at significantly higher prices.
$DRAM $SKHY $SNDK https://x.com/athuinvests/status/2105395525910745515
## @HunterAllen4 (THE GAP FATHER) · 09-30 16:17 · ♥30 ↻3 💬9 $CDNL
Another post-IPO drip salad that I think is being massively overlooked.
Remember $FPS. Another Gap father special is here. 😤
Don’t sleep. Don’t fade. Don’t miss.
I’ll be a bull on this bush until she grows leaves. High conviction.
If you haven’t heard of this one yet? Join through the link in bio. I’m researching 24/7 to find absolute bottoms.
The Gap Father Swing Channel stays HOT, and some of the rip salads loading in there are absolutely worth looking into.
We got this one into Telegram about 6% ago.
Yes, it could still dip toward $23, but the setup here is getting too interesting to ignore.
Cardinal Infrastructure Group is a Raleigh-based, vertically integrated civil/site contractor covering wet utilities, grading, blasting, paving and full site development across the Southeast. NC, SC and now Atlanta.
It IPO’d in December 2025 and is essentially a regional infrastructure roll-up riding population growth, housing, manufacturing reshoring, DOT spending and the early wave of data-center site development.
The growth is nuts. 2025 revenue was ~$456M, up 45%. Q1 2026 hit $168M, up 105%, followed by a record $226.9M in Q2, up 114%.
H1 revenue reached ~$394M, up 110%. Management has now raised full-year revenue guidance twice to $880–900M, roughly 95% growth at the midpoint.
And the hidden number I really care about: organic growth is still enormous. Q1 was ~64% organic and Q2 was roughly 56–64%. This isn’t simply a company buying revenue.
Backlog sits around $866M, up 35% YoY, with ~80% repeat customers. Adjusted EBITDA was $28.1M in Q2 and $54.9M for H1.
Margins compressed because of weather, subcontracted labor and investments required to scale into new markets, but management still expects 16–18% adjusted EBITDA margins for 2026, with a longer-term target in the low-20s as more work becomes self-performed.
Then there’s the acquisition machine.
CDNL has completed 9 deals since 2021, including ALGC in Atlanta and Piedmont Pipe, with Allied Paving expected to close in early October.
Allied brings roughly $108M of revenue and ~20% adjusted EBITDA margins while adding in-house paving capacity around Atlanta.
That’s the flywheel: buy local operators → deploy crews across the network → bring more work in-house → bid larger projects → keep consolidating the Southeast.
And yes, the data-center angle is real just don’t confuse CDNL with $FPS or $LGN.
CDNL is the dirt-and-pipe layer.
It already landed a $24M Phase 1 award for a multi-phase mission-critical campus, with full self-performed civil scope and substantial completion expected in 2027. Management says bidding activity is strong across NC and GA, while remaining disciplined about margins.
So why did the stock get obliterated?
It ran from the IPO into the mid-$90s, then gave back roughly 60–70% after Q2 despite a massive revenue beat and another guidance raise. The market focused on the EPS miss and margin compression.
Now we’re sitting around the $22–28 post-IPO base, with RSI around 33 and momentum still beat up.
That’s exactly why I’m watching it.
If $CDNL stabilizes margins while continuing to convert backlog and win larger industrial/data-center site work, I think the market has plenty of room to rethink the valuation.
Obviously, construction is lumpy. Weather, labor, housing, geographic concentration, integration and post-IPO volatility are real risks. If the margin recovery doesn’t happen, the market may continue treating this like a volatile regional contractor.
But after the collapse?
This bush is looking awfully interesting. 🌱
Rip salad is brewing. https://x.com/HunterAllen4/status/2105331196645241125