Cloud hosting that scales with the traffic, not against a cap
Most hosting gives you a slice of one server and throttles you at the edge of it. This platform spreads your site across a cluster and lets it draw what it needs during a spike, with no per-account CPU, memory or process limits.
- Autoscaling
- Capacity follows demand
- High availability
- Load balanced, self-healing
- Free
- Migration and global CDN
- 100% uptime
- Measured over the last twelve months
Your site is not pinned to one server
This is the part that matters most. A busy hour does not queue behind a resource limit, because there is no limit drawn around your account.
Each block is a unit of capacity. The grey blocks are what a fixed plan gives you all day, whether you need them or not.
Capacity follows demand, automatically
When requests climb, processing power, memory and bandwidth are allocated from across the platform within seconds. When the spike passes, they go back. Nothing to provision, nothing to upgrade mid-campaign.
Busy sites are moved, not throttled
A site under sustained load is shifted onto dedicated capacity so it keeps answering, and so nobody else on the platform feels it.
The whole platform is the resource pool
You are not buying a fixed slice of one machine. Your site can reach across many servers, which is what makes a traffic surge survivable rather than fatal.
Campaign launches, a television mention, Black Friday, a post that travels further than expected: these are the moments hosting is actually judged on, and they are the moments a fixed resource cap hurts most.
Four sizes, one platform
Every plan runs on the same autoscaling cluster with the same absence of resource caps. What changes is the number of sites, the storage and what is included before you add anything.
Launch
A business site or a handful of small ones, on the same platform the big plans use.
- Websites1
- NVMe storage25 GB
- BandwidthUnmetered
- CPU and memoryNo caps
- Mailboxes10 × 10 GB
- DatabasesUnlimited
- Autoscaling and high availability
- Free global CDN
- Free wildcard SSL, renewed automatically
- Unlimited free migrations
Business
The tier most businesses settle on: room for a portfolio of sites, with backups already included.
- Websites5
- NVMe storage50 GB
- BandwidthUnmetered
- CPU and memoryNo caps
- Mailboxes25 × 10 GB
- DatabasesUnlimited
- Backups included, 30 days of snapshots
- Staging sites and one-click cloning
- Free global CDN and wildcard SSL
- Free migrations, any number
Scale
For stores and applications where an hour of slowness is measured in lost orders.
- Websites25
- NVMe storage100 GB
- BandwidthUnmetered
- CPU and memoryNo caps
- Mailboxes100 × 10 GB
- DatabasesUnlimited
- Object cache for stores and heavy applications
- Priority queue on chat and tickets
- Quarterly performance review with an engineer
Agency
One account, Unlimited client sites.
- WebsitesUnlimited
- NVMe storageUnlimited GB
- BandwidthUnmetered
- CPU and memoryNo caps
- MailboxesUnlimited × 10 GB
- DatabasesUnlimited
- Backups included, 30 days of snapshots
- Named engineer for migrations
- One consolidated monthly invoice
Longer terms lock the price for the whole term. Site counts and storage are the only ceilings; there is no cap on CPU, memory, processes or bandwidth on any plan. Move up or down whenever you like and we prorate the difference.
Three things you can bolt on
All three are optional, all three are billed with the plan, and any of them can be switched off again.
Turbo
The acceleration layer: compute nodes running at up to 4.20 GHz, your site pre-fetched by every CDN location worldwide, and a 5 GB database quota across every database on the package.
- Processors up to 4.20 GHz for requests that cannot be cached
- Pre-fetched by every edge, not cached on first request
- 5 GB database quota, applied to every database
- One click on or off, with a cache report
Backups
Daily protection for sites that change occasionally, with a dated timeline you restore from yourself.
- 30 days of webspace snapshots
- 30 days of database snapshots
- Restore individual files and folders
Backups Pro
Greater retention, email backups and stronger recovery coverage, for sites that change every day.
- 30 days of webspace snapshots
- 60 days of database snapshots
- 30 days of email snapshots
- Every email account backed up
- Restore individual files and folders
Built so one failure is not your outage
Autoscaling and high availability are the same architecture seen from two angles: many servers behind one site.
One node failing changes the picture for us, not for your visitors.
Load balanced across nodes
Requests are distributed across multiple web nodes. A node that stops answering is taken out of the pool automatically and the others carry the traffic.
Redundant at every layer
Web, database, mail and storage each run as separate redundant services, so pressure or failure in one does not cascade into the others.
Storage that tolerates failure
Files sit on replicated NVMe arrays built to survive drive and server failures without taking sites offline.
Watched continuously
Health checks run constantly and our engineers are alerted on failures. Most incidents are handled before anyone notices enough to open a ticket.
Every plan carries a 99.9% uptime commitment, and the platform measured 100% over the last twelve months.
No LVE limits, and that changes everything
Most shared and cloud hosting in this market runs on a container system that caps each account's CPU, memory, processes and I/O. Hit the cap and your visitors wait, even though the server has capacity sitting idle.
Hosting with per-account limits
- CPU and memory capped per account, whatever the server has spare
- Requests queue once the cap is reached, so the site feels slow at exactly the wrong moment
- Entry processes and I/O throttled, which hits checkout and admin first
- The fix offered is always a bigger plan, bought under pressure
- Resource warnings that mean nothing to your customers
This platform
- No per-account CPU, memory, process or I/O caps
- Spikes are absorbed by adding capacity, not by throttling you
- Sustained load moves your site onto dedicated capacity automatically
- Unmetered bandwidth on every plan
- You are sized by what you host, not by what you might consume in an hour
Plans differ by how many sites and how much storage you need, not by how hard you are allowed to work the servers.
Two things most hosts leave as they found them
Off-the-shelf hosting software is built for one server. Running it across a cluster means changing it, and these are the two places where that work shows up in what you get.
A PHP stack tuned for a cluster, not a box
The standard speed tools assume your site lives on a single machine, so they scale badly the moment it does not. The platform runs a customised PHP-FPM and opcode cache instead, written to hand a site the resources it needs while it is busy rather than rationing them.
- Every version from PHP 5.6 upwards runs faster for it
- The gain is larger on recent versions of PHP
- No resource-limiting layer sitting between your site and the hardware
- Which is why there are no caps on cores, memory, bandwidth or processes
A stock configuration with a resource limiter on top, which is what produces the per-account CPU and memory ceilings you are used to seeing.
Balancing for load, not only for failure
Almost every host balances for failover: a server dies and your site is moved so it stays online. That is disaster recovery, and it is here too. The difference is what happens on the far more common day when nothing fails and everything is simply busy.
- Monitors watch the actual load and move traffic before it queues
- Traffic spreads across many servers rather than waiting on one
- A failed node is still taken out of rotation automatically
- Redundant hardware and a 10 Gb network underneath both behaviours
Failover only. Your site stays online when a server dies, but stays slow when a server is merely overloaded, which happens far more often.
Both of these are the reason the no-limits claim on this page is possible at all. You cannot remove the caps without first removing what made them necessary.
What Turbo actually changes
Autoscaling decides how much capacity answers your traffic. Turbo changes three separate things: where your code runs, how far ahead the cache is warmed, and how much room the database has.
Dynamic work runs on standard nodes, and the first visitor in each region waits while the cache fills.
The page is already at every edge before anyone asks for it, and what cannot be cached runs on faster processors.
Illustrative, not a benchmark: the gain depends on how cacheable your pages are and how much of your traffic is dynamic. We will measure yours before and after if you ask.
Processors up to 4.20 GHz
Requests are routed to high-frequency compute nodes running at up to 4.20 GHz. It shows up most in the work that cannot be cached at all: admin screens, checkouts, search and imports.
Pre-fetched by every edge
Your site is pulled into every CDN location worldwide in advance, instead of being cached the first time somebody asks for it. The first visitor in a region gets the fast response rather than paying for it.
5 GB of database space
The database quota rises to 5 GB, and it applies to every database on the package rather than to one of them. That ceiling is usually what constrains a large catalogue or a busy membership site.
On and off in one click
Turbo goes on without downtime and comes off the same way. A cache report in the control panel shows you what is being served from cache and what is still reaching your application.
Included, unlimited, on every plan
Images, scripts and stylesheets are served from edge caches close to the visitor instead of travelling back to the origin every time. It is part of the platform rather than a paid extra, and it is on for every plan including the smallest.
Anycast routing
Visitors are answered by the nearest healthy edge automatically, with no per-region configuration to maintain.
Unlimited traffic
No bandwidth allowance to watch and no overage invoice after a busy month.
Works with the cache
The CDN and the page cache purge together, so a publish is live everywhere rather than in some places.
Better Core Web Vitals
Shorter round trips improve the loading metrics search engines measure, particularly for visitors furthest from the origin.
Edge locations are part of the platform and change as it grows. Ask us which edges serve your main audience and we will tell you what we are seeing.
The full list, without an upsell
All of this is on every plan, Launch included. Turbo and the two backup tiers are the only paid extras.
Performance
- No CPU, memory or process caps
- NVMe storage throughout
- Free global CDN, unlimited traffic
- Tuned PHP-FPM with OPcache
- PHP 5.6 through 8.4, per site
- HTTP/2 and IPv6
- Object cache for dynamic sites
- Unmetered bandwidth
Security
- Free wildcard SSL, auto-renewed
- Web application firewall
- Layer 3, 4 and 7 DDoS protection
- Daily malware scanning
- Brute-force and login protection
- Two-factor authentication
- IP and country blocking
- Hotlink protection and permissions checker
Developer
- SSH access with two-factor
- Git version control
- SFTP and FTP
- Scheduled tasks (cron)
- One-click installers for 70+ applications
- WordPress staging and cloning
- WP-CLI on WordPress sites
- Access and error logs
Email, DNS and domains
- 10 GB mailboxes on your own domain
- Unlimited forwarders and autoresponders
- Webmail, IMAP, POP and mobile sync
- Spam and virus filtering
- Free mail migration
- Full DNS record editing
- One-click records for Google Workspace or Microsoft 365
- Unlimited MySQL databases
Why this is not a shared server with a nicer name
Cloud hosting only means something if a failure somewhere does not become an outage for you, and if a busy hour does not become a throttle.
Requests arrive at the platform rather than at a machine. Routing picks capacity that is healthy and free, the autoscaler adds more when demand climbs, and anything that stops answering is taken out of rotation.
Web nodes
Many servers answer for one site, added and removed as demand moves through the day.
Database cluster
Replicated, with its own failover, so your queries do not queue behind a neighbour's.
Mail platform
A separate service on separate machines, so a busy website never holds up your email.
Replicated storage
Redundant NVMe arrays that tolerate a drive, or a whole server, failing without losing anything.
Many servers, one site
Your site runs across the platform rather than on a machine you could point at. That is what makes both the scaling and the failover possible.
Separate layers
Web, database, mail and control panel run as independent redundant services, so pressure on one does not drag the others down.
Isolated neighbours
A site under heavy load is moved onto dedicated capacity, so your performance does not depend on somebody else's bad day.
Monitored around the clock
Health checks and alerting run continuously, with engineers on call rather than an answering service.
Scale without rebuilding
Changing plan changes the ceilings in place: same account, same IP, no reinstall and no migration window.
Hosted on our platform with local peering for Pakistani visitors and edge caches worldwide for everybody else.
Everything side by side
Add-on means available at the price above; included means it is already in the plan.
| Feature | Launch | Business | Scale | Agency |
|---|---|---|---|---|
| Websites | 10 | 25 | 100 | Unlimited |
| NVMe storage | 50 GB | 150 GB | 400 GB | 1 TB |
| Bandwidth | Unmetered | Unmetered | Unmetered | Unmetered |
| CPU, memory, processes | No caps | No caps | No caps | No caps |
| Autoscaling | ||||
| High availability | ||||
| Global CDN | ||||
| Mailboxes (10 GB each) | 10 | Unlimited | Unlimited | Unlimited |
| MySQL databases | Unlimited | Unlimited | Unlimited | Unlimited |
| Wildcard SSL | ||||
| Web application firewall | ||||
| DDoS protection | ||||
| Malware scanning | ||||
| Backups included | ADD-ON | |||
| Snapshot retention | 30 days | 30 days | 30 days | 30 days |
| Email backups | ADD-ON | ADD-ON | ADD-ON | ADD-ON |
| Turbo acceleration | ADD-ON | ADD-ON | ADD-ON | ADD-ON |
| Staging and cloning | ||||
| Object cache | ||||
| SSH, Git and WP-CLI | ||||
| Separate client logins | ||||
| Free migrations | Unlimited | Unlimited | Unlimited | Unlimited |
| Support | 24/7 | 24/7 | 24/7 priority | 24/7 priority |
| Uptime commitment | 99.9% | 99.9% | 99.9% | 99.9% |
Move between plans at any time. We prorate the difference and nothing needs rebuilding, because the ceilings change rather than the machine.
We move it, you check it
Unlimited free migrations on every plan, from any control panel or a plain backup file.
- Files and databases copied
- Mailboxes moved with their contents
- DNS records rebuilt and tested
- Cut over with no downtime
No data loss, no downtime window to book, and nothing for you to reinstall.
Send us access
Control panel login, or an archive if that is easier. We work the rest out from what is there.
We copy and test
Files, databases, mailboxes and DNS are copied and tested on a staging URL before anything changes for your visitors.
We cut over
Usually overnight. DNS switched, mail verified, and we stay on the ticket until you confirm it is right.
What each plan tends to run
A rough guide rather than a rule; tell us the traffic and we will size it properly.
Business websites
Brochure sites, service businesses and anything where the enquiry form is the revenue.
WordPress and WooCommerce
Caching, object cache and staging tuned for the stack most Pakistani stores run on.
Applications
Laravel, Node and custom software that needs guaranteed CPU, SSH and deployment from Git.
Agencies
Many client sites on one account, each isolated, with one invoice and separate logins.
Related pages
Website Turbo
What the Turbo add-on actually changes: faster processors, pre-fetching to every edge and a larger database quota.
Read more →Register a domain
Search and register the name, with DNS, transfer lock and reminders included.
Read more →Transfer a domain
Move an existing domain across. It adds a year and nothing goes offline.
Read more →Domain pricing
Registration, renewal and transfer prices for every extension, in rupees or dollars.
Read more →Before you choose a plan
What people ask us about autoscaling, limits and pricing.
What does autoscaling actually mean here?
When demand rises, the platform allocates more processing power, memory and bandwidth to your site from across many servers, then releases it when the spike passes. You do not provision anything, and there is no upgrade to buy in the middle of a busy afternoon.
What does autoscaling actually mean here?
When demand rises, the platform allocates more processing power, memory and bandwidth to your site from across many servers, then releases it when the spike passes. You do not provision anything, and there is no upgrade to buy in the middle of a busy afternoon.
Move in this week
Tell us what you run today. We will size the plan, migrate it free and stay on the ticket until it is stable.