Solutions
A second home for the day one big cloud goes down.
Founders and ops leads at 5 to 50 person software companies who picked a readable bill over a big name, and know one bad week of downtime costs more than a year of savings.
Why now
One region took the internet down for a night.
One region took the internet down
AWS’s own post-event summary records increased DynamoDB error rates in us-east-1 from 11:48 PM PDT on 19 October to 2:40 AM PDT on 20 October 2025, with EC2 launches, Lambda, Fargate and EKS impaired after it. [1]
SourceAmazon Web Services post-event summary, October 19 to 20, 2025. The full citation and link are in the footer notes.
What it means for you
The largest cloud lost a region for a night and small companies were simply off. If you have one cloud, you have one point of failure.
What waiting costs
A small company has no second cloud and no team to fail over by hand. A night offline is a payroll you cannot run, orders you cannot take, and customers who try the competitor.
Why Triple X
Triple X gives a small team a second home priced by the hour, with a free tier to start, standard tools, and failover inside your own private cloud and across the network if you allow it.
In plain English
The big cloud will go down again, and you will not be the first call it returns. For the price of a few machines a month, keep a copy of what matters somewhere you can switch to.
What you get
Four things, each a product you can open.
Failover you can test
Launch two nodes, stop one on purpose, watch the work come back. 1,800+ nodes behind your private cloud.
OpenA monthly number you can read
Hourly, published, transfer included in the math, nothing to leave.
OpenYour tools unchanged
Standard machines, Kubernetes, the S3 API, PostgreSQL. Your scripts work.
OpenA warm copy, paid by the hour
Keep staging or a copy of production here and pay only while it runs.
Open
Which way in
Start where it fits. Move later without a new vendor.
Triple X Cloud nodes
No hardware of your own. Join the waitlist and start when your seat opens.
Your own devices
Triple X connects your racks or devices to a private cloud of your own.
Both
Sensitive work on your devices, the rest on ours, one private cloud.
Objections you will hear
What someone in the room will say.
My current provider is already cheap.
Tier D runs at 70 against hyperscaler list and Tier B at 90, with outbound from $0.01 per GB and nothing to leave. Compare the monthly number, not the headline.
Distributed sounds fragile.
There is no single building, no single switch, no single region to fail. When one node goes, the work moves.
I do not want to learn something new.
You will not. Standard machines, standard Kubernetes, the S3 API, PostgreSQL.
What if you go away?
Your data is on the S3 API and your clusters are standard. No exit fee, in the contract. You can leave in an afternoon.
What yes looks like
The list a buyer brings to the table.
- A monthly number at or below today’s bill, transfer included
- Failover they can test by stopping a node
- Standard tools, nothing proprietary
- No exit fee and no minimum term
- A human to call
- Capacity that grows without anyone building a data centre
Get started
Move staging first.
Then stop a node on purpose.
- 01
Join the waitlist.
Tell us what you want to run and where. No credit card.
TakesTwo minutes
- 02
Run the pilot.
Ten nodes, 60 days, a pass mark you set in writing.
Takes60 days
- 03
Decide on day 60.
Extend, or leave at the transfer rate and nothing else.
WhoYou
We never ask for your root credentials, API keys or bank details by email.
The first 60 days
A pilot with a pass mark you set.
Move staging, pull a plug on purpose, and read one invoice beside the current one before production moves.
Week 1
Staging and a warm copy
Create the account on the free tier, move staging, and stand up a copy of the production database. Set a budget with an alert.
Weeks 2 to 4
Stop a node in business hours
Stop a node on purpose and time the recovery. Restore the database to a point in time. The pass mark is the founder's: recovery time, restore time, and a monthly number at or below today's bill.
Day 30
Two invoices side by side
Read the first invoice beside the current provider's for the same workloads, transfer included.
Day 60
Decision
Move production, keep the warm copy only, or stop. Leaving costs the transfer rate and takes an afternoon.
Questions
What buyers ask before they sign.
In the order they come up. Every answer ends with where to check it.
10 answers
All questionsCan we move staging this week without learning a new stack?
Yes. Standard machines, standard Kubernetes with 428 of 428 conformance tests passed, the S3 API and PostgreSQL 17. Your scripts work. The free tier needs no credit card and is enough for the first machine.
Proofthe conformance run, with the CNCF listing in progress, 2026.[2]
What will the monthly number be?
A small machine is $0.024 an hour, about $17.50 a month. The default two-node cluster is about $35. Object storage is $0.02 per GB; outbound from $0.01 per GB. Databases are quoted per size. Set a budget with an alert before anyone else does.
Proofthe published rate card and the budget controls, August 1, 2026.[3]
Is it cheaper than what we pay now?
Tier D is indexed at 70 and Tier B at 90 against public cloud list price, modelled in July 2026, not measured from migrations. Compare the monthly number with transfer included, not the headline rate.
Proofthe tier table and the price index note, July 2026.[4]
What happens when a node dies at two in the morning?
The work moves to another node inside your private cloud, then across 1,800+ nodes if you allow it. Tier B commits to 99.95%, about 22 minutes a month. There is no single building, switch or region to fail. Test it by stopping a node yourself.
Proofthe fleet count, company-stated, October 2026, and the Reliability page.[5]
Can we keep a warm copy of production here and pay only while it runs?
Yes, by the hour. The managed database runs three nodes, keeps seven days of point-in-time history and stores backups in a separate region. Stop the copy and the meter stops with it.
Proofthe managed databases overview in the documentation, August 1, 2026.[6]
Who can see our customers' data?
You and the people you add to your private cloud. The platform underneath sees that a workload exists, how big it is and what it costs, not what is in it, and its operator actions are logged.
Proofthe access table on the Security page and the SOC 2 Type II report under NDA.[7]
What if Triple X goes away, or we want out?
Your data is behind the S3 API and your clusters are standard, so you can leave in an afternoon. No exit fee and no minimum term, in the contract. Leaving costs the outbound transfer rate.
Proofthe no-minimum-term rule in the published pricing and the no-exit-fee clause in the contract.[3]
What do we tell a customer who asks where we host?
That the platform underneath holds a SOC 2 Type II attestation with an unqualified opinion, available under NDA, and that the uptime commitment is written in the master services agreement by tier. Point them to the Security page.
Proofthe SOC 2 Type II report under NDA.[7]
We have two servers in a cupboard. Can they count?
Yes. They join your private cloud as your own devices, with the same console and tooling, and Triple X quotes the platform charge in writing. Most small companies start on hosted nodes and keep the cupboard for the warm copy.
Proofthe Sovereign page and a written quote from Triple X.
Is there a human to call, and what is not automatic?
Write through the contact page; security reports are acknowledged within two business days. Limit increases are reviewed in 1 to 2 business days. Nobody credits you automatically for downtime: the claim window is in the agreement, so file the claim.
Proofthe Contact page and the credit mechanism on the Reliability page.
No questions under this term yet. Ask Triple X.