Twice the RAM per core. Same managed stack.
Memory-hungry workloads need memory headroom. Shared CPU VPS plans give you 4 GiB DDR5 per vCPU, double the ratio of Basic, with the same shared CPU economics. Seven plans from 4 GiB to 64 GiB RAM, fully managed since 2001. Free migration from your current host.
Shared CPU VPS, explained
A Shared CPU VPS is a virtual private server where the CPU cores are scheduled across multiple tenants, but memory and storage are 100% allocated to you. Shared CPU Cloud Cubes go further: you get 4 GiB DDR5 RAM per vCPU, double the ratio of an entry-level managed VPS, with the same shared CPU economics.
Most managed VPS plans give you a 2:1 RAM-to-CPU ratio. That works for low-traffic WordPress and brochure sites, but breaks down quickly for memory-hungry workloads: large WooCommerce stores, multi-database stacks, in-memory caches, and PHP-FPM pools serving real concurrency. When your bottleneck shifts from CPU to RAM, paying for more CPU you do not need is wasteful. A Shared CPU VPS solves this by giving you the memory headroom without the dedicated-core price tag.
You still get every operational feature: 24/7 monitoring, security patching, performance tuning, daily backups, mail gateway, free migration. The CPU model is the only thing shared. Shared CPU Cloud Cubes are the right fit when your workload is RAM-bound, not CPU-bound, and you want to grow past 16 GiB RAM without jumping straight to dedicated cores.
Shared CPU Cloud Cubes at a glance
Pricing
Shared CPU Cloud Cubes start at $18/month with annual billing ($22 monthly), going up to $293/month for the 64 GiB plan. Every plan includes 100% management, daily backups, free migration, and 500% SLA. No setup fees, no contracts, cancel anytime.
billed monthly
billed $432/year
billed monthly
billed $876/year
billed monthly
billed $1,320/year
billed monthly
billed $1,752/year
| Plan | Cores | Memory | Storage | Transfer | Management | Price / mo | |
|---|---|---|---|---|---|---|---|
| Shared 4GB | 1 vCPU +3.00 GHz | 4 GiB DDR5 | 50 GB NVMe + 100 GB free S3 | 5 TB 50 Gbps DDoS | 100% Fully Managed 500% SLA | $22/mo$264/year $18/mo$216/year | Order Now Order Now |
| Shared 8GB | 2 vCPU +3.00 GHz | 8 GiB DDR5 | 100 GB NVMe + 200 GB free S3 | 10 TB 50 Gbps DDoS | 100% Fully Managed 500% SLA | $44/mo$528/year $36/mo$432/year | Order Now Order Now |
| Shared 16GB | 4 vCPU +3.00 GHz | 16 GiB DDR5 | 200 GB NVMe + 400 GB free S3 | 20 TB 50 Gbps DDoS | 100% Fully Managed 500% SLA | $88/mo$1,056/year $73/mo$876/year | Order Now Order Now |
| Shared 24GB | 6 vCPU +3.00 GHz | 24 GiB DDR5 | 300 GB NVMe + 600 GB free S3 | 30 TB 50 Gbps DDoS | 100% Fully Managed 500% SLA | $132/mo$1,584/year $110/mo$1,320/year | Order Now Order Now |
| Shared 32GB | 8 vCPU +3.00 GHz | 32 GiB DDR5 | 400 GB NVMe + 800 GB free S3 | 40 TB 50 Gbps DDoS | 100% Fully Managed 500% SLA | $176/mo$2,112/year $146/mo$1,752/year | Order Now Order Now |
| Shared 48GB | 12 vCPU +3.00 GHz | 48 GiB DDR5 | 600 GB NVMe + 1.2 TB free S3 | 60 TB 50 Gbps DDoS | 100% Fully Managed 500% SLA | $264/mo$3,168/year $220/mo$2,640/year | Order Now Order Now |
| Shared 64GB | 16 vCPU +3.00 GHz | 64 GiB DDR5 | 800 GB NVMe + 1.6 TB free S3 | 80 TB 50 Gbps DDoS | 100% Fully Managed 500% SLA | $352/mo$4,224/year $293/mo$3,516/year | Order Now Order Now |
Annual billing gets you 2 months free per year (16.67% off the monthly rate). Software licenses (cPanel, LiteSpeed, Imunify360) are billed at full price and are not discounted by billing cycle.
Match your workload to the ratio
Shared CPU Cloud Cubes work best when your bottleneck is RAM, not CPU. Below are the four most common workloads where the 4 GiB DDR5 per vCPU ratio actually matters. If your situation matches one of these, Shared CPU is likely the right tier.
You run WooCommerce or Shopify-equivalent platforms with real catalogs, real inventory, and real concurrent shoppers. Memory headroom keeps checkout response times flat under load while object caching scales without thrashing.
Recommended size: Shared 16GB or 24GB
You run several databases on one server (MySQL plus Postgres, Redis, Elasticsearch). Each engine wants memory and they fight when total RAM is constrained. Shared CPU gives every database room to breathe without paying for dedicated cores.
Recommended size: Shared 24GB or 32GB
You run 10+ WordPress sites or a WordPress Multisite network with 50,000+ monthly visitors. PHP-FPM pools eat memory fast and shared 2:1 ratios choke. The 4:1 RAM-to-CPU model handles the load without overpaying for compute.
Recommended size: Shared 16GB or 32GB
Your SaaS has product-market fit, 5,000 to 50,000 active users, and is hitting memory limits on entry-level VPS plans. Queue workers, background jobs, and in-memory state need RAM that does not exist on 2:1 plans.
Recommended size: Shared 24GB or 48GB
Picking the right tier
Shared CPU sits between Basic and Dedicated for a reason: it gives you the memory headroom of a bigger server with the shared CPU economics that keep prices honest. Here is what that means in practice, and where Shared CPU genuinely fits versus the tiers above and below.
Basic Cloud Cubes ship with a 2:1 RAM-to-CPU ratio: 2 GiB of memory per vCPU. That is the right ratio for low-traffic WordPress, brochure sites, and dev environments where the workload is light. It stops being the right ratio the moment you load up object caches, run multiple databases, or serve real concurrency through PHP-FPM.
Shared CPU Cloud Cubes give you 4 GiB DDR5 RAM per vCPU, double the ratio of Basic, with the same shared CPU pricing model. Memory is yours, NVMe disk is yours, I/O queue is yours. Only the CPU cores are scheduled across tenants, and that is what lets us price 16 GiB of RAM at $73/month annual instead of $146/month on Dedicated.
You also get the same operations stack: 24/7 monitoring, security patching, daily backups, mail gateway, free migration. You access the server through RemarkablePanel, run multiple PHP versions, configure databases the way you need them. Nothing about the management changes between tiers, only the resource ratios.
4:1
RAM-to-CPU Ratio
4 GiB DDR5 per vCPU. Double Basic's 2:1 ratio. Memory headroom for serious workloads.
100%
Dedicated Memory
Your RAM stays yours. No memory ballooning, no swap surprises from neighbors.
64 GiB
Top-end RAM
Scale to 64 GiB without jumping to dedicated cores. Basic stops at 16 GiB.
500%
Uptime SLA
5x downtime credited back, not the 1x most hosts offer.
Shared CPU is the right tier when your workload is RAM-bound rather than CPU-bound, and you want to grow past 16 GiB without paying for dedicated cores you do not need.
Shared CPU is the value sweet spot when memory is your real constraint and CPU is not.
The wrong tier hurts you twice: you pay for resources you cannot use, and you do not get the characteristics you actually need. Be honest about your workload before you order.
We tell you this because we would rather you land on the correct tier from day one. Tell us your workload and we will recommend a tier.
Choosing the right tier
Cloud Cubes come in four tiers, each tuned for a different workload shape. The CPU model decides between the first three, and storage capacity decides whether the fourth is the right answer. Here is the honest breakdown of when each tier is the right fit.
From $8/mo · 2 to 16 GiB
Move down to Basic if
From $18/mo · 4 to 64 GiB
Stay on Shared CPU if
From $40/mo · 8 to 128 GiB
Step up to Dedicated CPU if
From $46/mo · 8 to 128 GiB
Look at Storage Optimized if
The boundary
"Managed hosting" is overused. Some hosts call themselves managed and still leave OS patches, mail config, and security alerts on your plate. Here is exactly what we handle, what stays under your control, and where the line is. No ambiguity.
Server administration, security, monitoring, and operations are our job, included in your plan.
Your business, your code, your data. We do not touch these without your explicit authorization.
No surprise support tickets, no scope creep. These are deliberately outside the managed scope.
Question we did not cover?
Talk to a humanThis is what "managed" should mean. Compare it to providers who call themselves "managed" while leaving security patches as your job. Those are not actually managed.
Included free with every Cube
On most managed VPS hosts, your server runs the control panel, DNS server, and backup engine alongside your actual workload, eating CPU, RAM, and disk before you host a single site. RemarkablePanel moves all three to our shared infrastructure at no charge. Your VPS runs only what matters: web, mail, and databases.
Traditional VPS
Your Cloud Cube
The web UI, account database, billing integration, and admin services all run on our shared cluster. You get a full-featured panel without it consuming a single MB of your VPS RAM or a single percent of your CPU. cPanel-style power without the cPanel-style overhead.
Authoritative DNS for every domain on your account runs on our geographically distributed name servers. Faster lookups for visitors worldwide, automatic failover if a region goes down, no DNS daemon eating memory on your server, and no DNS hijacking risk if your VPS is ever compromised.
Daily snapshots and off-site backups run on dedicated backup hardware, not your VPS. No CPU spikes during backup windows, no disk I/O contention slowing your site, and no backup-induced load average alerts at 3am. Restores happen the same way: from our infrastructure, not yours.
The category question
Before choosing a tier, choose a category. These three hosting types are not really competitors, they solve different problems for different buyers. Pick the wrong category and tier choice will not save you.
| Feature | Shared Hosting | Managed VPS Cloud Cubes | Unmanaged Cloud DigitalOcean / Linode / Vultr |
|---|---|---|---|
| Resources | Shared with hundreds of others | 100% allocated to you | 100% allocated to you |
| Server administration | Provider-managed | Provider-managed | Your responsibility |
| Security patching | Provider | Provider | Your responsibility |
| Monitoring and alerts | Basic uptime only | 24/7 proactive | DIY (you set it up) |
| Performance under load | Throttled at peak | Consistent | Consistent (if you tune it) |
| Custom software | Limited | Most stacks supported | Anything (you install it) |
| Root access | No | Available on request | Yes (default) |
| Mail gateway and antispam | Basic, shared | Rspamd, dedicated | You configure it |
| Migration help | Rarely free | Free, full handoff | Not included |
| Best for | Personal blogs, simple sites | Agencies, eCommerce, SaaS, growing businesses | Teams with DevOps capacity |
| Typical cost | $3 to $15/month | From $8/mo (Basic, annual) | $5 to $20/month + your time |
A personal blog, a brochure site, a hobby project. Most CloudCubes customers came from shared hosting after outgrowing it. If your site is still small and growing slowly, shared hosting is a fine starting point.
This is where most growing businesses land. Dedicated RAM, dedicated storage, predictable performance, full operations included. The right answer for agencies, eCommerce stores, SaaS in early traction, and serious WordPress workloads.
DigitalOcean, Linode, and Vultr are excellent if you have the operations capacity to run them. The dollar savings vs managed are real, but the time cost is real too. Make sure that math works for you before signing up.
Pricing transparency
The honest answer: most price differences come down to what is actually included. A $20 droplet and a $73 managed VPS at 16 GiB are not the same product, and comparing the sticker prices is misleading. Here is what each price tag actually covers, with no asterisks.
Listed price is the total. No upgrades to read your email, no fee to talk to a human, no surprises on month two.
DigitalOcean, Linode, and Vultr advertise from $20/mo for similar memory tiers. By the time you have a real production setup, that number is much higher.
Honest disclosure: a few things are not in the base price. Most customers do not need them, but you should know what they are.
Have a specific cost question?
Ask before you commitThe honest comparison is not "$73/mo Shared CPU 16GB vs $20/mo unmanaged droplet." It is "$73/mo with operations included" versus "$20/mo plus $50 to $150/mo in management add-ons or your own DevOps time." When you do the actual math, managed always wins below the threshold of having a full-time sysadmin on payroll.
The RC difference
Every plan ships with the full stack. No tiers, no add-ons required, no surprise charges. Twelve operational features that most hosts charge extra for, all included in your base price.
Every plan, every tier, no exceptionsServer administration, security hardening, OS patches, performance tuning, incident response. Real engineers since 2001, not chatbots, not outsourced ticket farms.
25 years of operations
We watch CPU, RAM, disk, network, and service-level signals around the clock. If something breaks, we act before you notice. No support ticket required.
Acted on, not just measured
Automatic block-level snapshots taken every 24 hours, retained 7 days. Restore to any point in the last week on request, no extra charge.
7-day retention, instant restore
Network-edge mitigation up to 50 Gbps included. Automatic null-routing for sustained attacks. No additional fees, no scrubbing center surprises.
50 Gbps, automatic
Dedicated Rspamd-based mail gateway filters incoming spam before it hits your server, and protects your sender reputation on outbound. Almost no other VPS host bundles this.
Rspamd cluster, included
Off-site replication runs nightly to dedicated backup hardware. Zero impact on your Cube's CPU or disk I/O, and your data survives even if the server room burns down.
Off-site, dedicated infrastructure
Every plan includes off-site S3 storage at 2x your disk size, no extra charge. Use it for backup archives, large transfer staging, or external app data.
2x disk size, no add-on fee
All system, web, mail, and security logs aggregated and analyzed automatically. Anomalies trigger alerts, our team investigates, you stay informed.
Aggregated, monitored, escalated
We promise 99.99% uptime with a 5x credit if we miss it, not the standard 1x credit other hosts offer. One hour down equals five hours credited back, automatically.
5x downtime credit, automatic
Network-level firewall rules, intrusion detection, fail2ban, and SSH lockdown configured by us, maintained by us. You do not write iptables, ever.
Hardened by default, tuned by us
RemarkablePanel included on every plan, runs on shared infrastructure so it does not eat your Cube's resources. cPanel, DirectAdmin, and Plesk available with their own licenses.
RemarkablePanel, free
Switching feels like work. We do the work. Our team migrates your sites, databases, mail, DNS, and SSL certificates, plans your cutover, and handles the dirty parts. White-glove, included.
Free, white-glove, included
Verified reviews
Real feedback from real businesses. Independently verified on Google and Trustpilot, no edits, no cherry-picking.
After encountering numerous issues with my WordPress site while using other providers, a friend recommended RemarkableCloud to me. It's been two years since I made the switch, and I can confidently say I've never looked back. The server is not only fast but also incredibly stable. I haven't experienced any downtime in almost two years. The support team is impressively fast and knowledgeable, always ready to assist.
These people deserve nothing less than a 10/10. They're truly masters in the cloud field.
The tech support is excellent, and they offer very competitive pricing. They also migrated my cPanel VPS to a DirectAdmin VPS without noticeable downtime.
Problem resolution in no more than half an hour for simple issues, and within 4 hours for complex ones. Great service.
Excellent web hosting company. I really love all their products and services. Thank you so much.
Very professional. Fast, friendly, and consistent. Keep it up.
It's been a month since I signed up with RemarkableCloud.com to test their services and technical support. So far everything has been good. I'm happy with the service.
Most hosting SLAs only kick in after uptime drops below 99% or 99.9%, which is 7 to 87 hours of downtime per month before you see anything back. Ours starts from minute one, applied automatically, with no ticket required.
Credits start accumulating from the first minute of downtime. No minimum threshold, no waiting room, no fine print clauses about "scheduled maintenance" or "force majeure."
Credits are applied automatically to your next invoice. You do not need to open a support ticket, submit a claim form, or argue with billing about whether the downtime really happened.
While other hosts calculate fractions of a percent, we guarantee 5x compensation in service credit. It is not a marketing claim. It is written into the contract you sign when you order.
Typical hosting SLA
Credits only trigger after uptime drops below 99% or 99.9%. That is 7 to 87 hours of downtime per month before you see anything back, and the credit is usually 1x the lost time.
RemarkableCloud SLA
Credits from minute one. Five hours back for every hour down. Applied automatically, every time, with no action from you.
Memory headroom without paying for dedicated cores. Shared CPU Cloud Cubes start at $18/month with annual billing ($22 monthly), every operational feature included, no add-ons required.
No setup fees. No surprise charges. The price you see is the price you pay.
FAQ
Honest answers about Shared CPU Cloud Cubes specifically. If your question is not here, send it to us, we will answer it personally and add it to this list.
Shared CPU means your virtual cores are scheduled across multiple tenants on the underlying physical hardware. You get full access to your allocated vCPUs almost all the time, but the physical CPU cores are shared with other VMs. This is how DigitalOcean droplets, Linode shared instances, and most entry-level cloud servers work.
Memory and storage are NOT shared. Your RAM is dedicated to your VM, your NVMe disk space and I/O queue are dedicated to your VM. Only the CPU scheduling is shared, and that is what lets us price 16 GiB of RAM at $73/month annual instead of $146/month on a dedicated-core plan.
When your workload is RAM-bound rather than CPU-bound. The most common signs: PHP-FPM pools running out of memory under concurrent load, MySQL or Postgres database with a working set larger than your buffer pool, multiple memory-hungry services on one server (Redis + Elasticsearch + queue workers), or large WooCommerce stores with object caching.
If your CPU usage is under 50% but RAM stays at 80-95%, you do not need more cores, you need more RAM per core. Shared CPU's 4:1 ratio fixes that without paying for dedicated cores you are not using.
The CPU model is the same (shared vCPU). The RAM-to-CPU ratio is what changes. Basic gives 2 GiB RAM per vCPU. Shared CPU gives 4 GiB RAM per vCPU. Same operations stack, same management, same SLA.
Shared CPU also goes up to 64 GiB RAM per plan (Basic stops at 16 GiB), so it is the right tier for memory-hungry workloads that outgrow Basic. Storage and transfer scale proportionally with the larger memory tiers.
Under normal load, no. Shared CPU is designed for workloads that have CPU bursts but spend most time idle or in low utilization. WordPress, WooCommerce, multi-tenant CMS, and most application servers fit this profile because PHP/HTTP traffic is bursty by nature.
You will see throttling if you sustain 100% CPU usage for extended periods (think: video transcoding, ML inference, write-heavy OLTP databases, parallel batch processing). Those workloads need Dedicated CPU Cloud Cubes, where cores are not scheduled across tenants. We monitor your CPU patterns and will tell you if your workload is being throttled regularly.
Yes, and it is one of the most common use cases for this tier. WooCommerce is RAM-hungry: object cache, session data, cart state, product catalog, and PHP-FPM workers all live in memory. The 4:1 RAM ratio gives every layer breathing room.
For real volume stores (50+ orders per day, sustained concurrent shoppers), Shared 16GB or 24GB is the sweet spot. We tune Redis or Memcached for object caching, MySQL for write-heavy carts, and PHP-FPM pool sizes for your concurrency level. If your CPU stays consistently above 70% under normal traffic, we will recommend stepping up to Dedicated CPU.
Far more than Basic. As a rough guide: Shared 8GB handles 15-25 moderate WordPress sites, Shared 16GB handles 30-50, Shared 32GB handles 60-100. The 4:1 RAM ratio is what makes this scale work, since each PHP-FPM pool gets enough memory to handle its concurrency without thrashing.
If you run WordPress Multisite, Shared CPU is also a strong fit: a single Multisite network with 50,000+ monthly visitors fits comfortably on Shared 16GB or 32GB depending on plugin load.
When your workload is CPU-bound, not RAM-bound. Telltale signs: sustained CPU usage above 70-80%, p99 response times spiking under load while p50 stays flat, scheduled jobs taking longer than expected, or compliance requirements that prohibit shared scheduling.
Dedicated CPU Cloud Cubes give you cores that nobody else can touch, with the same management and SLA. They cost more per GiB of RAM than Shared CPU, but they deliver predictable performance under sustained load. We monitor your CPU patterns and will recommend the upgrade when we see it is needed.
Yes, every Shared CPU plan includes daily snapshots with 7-day retention plus off-site remote backups, no extra charge. Snapshots are block-level and fast to restore. Off-site backups run nightly to dedicated infrastructure so your data survives even if the server hardware fails.
You also get free S3 backup storage at 2x your disk size. If your Shared 16GB has 200 GB NVMe, you get 400 GB of free S3 storage included for archives or external backups.
Yes. Free migration is included on every Shared CPU plan, regardless of where you are coming from (cPanel host, Plesk host, WP Engine, Cloudways, Kinsta, AWS, DigitalOcean, anywhere). Our team handles the move: sites, databases, mail accounts, DNS records, SSL certificates, the whole stack.
We plan your cutover with you, do a test migration first, verify everything works at full traffic levels, then schedule the final DNS switch at a low-traffic time. Most customers experience zero downtime during migration.
We do not throttle, we do not bill you in surprise overage charges, and we do not suspend your service. Instead, we monitor your resource usage and reach out when you are consistently using 80% or more of your allocated RAM, CPU, or disk. We will recommend either an upgrade or a tuning pass to optimize your existing usage.
For network transfer, the included allowance is generous (5 to 80 TB per month depending on plan size) and most workloads never come close. If you do exceed it, we charge a small per-GB fee that we communicate before applying.
Yes, in either direction. Upgrades within the Shared CPU tier (4GB to 64GB) happen with zero downtime: we add the resources to your existing server in place. Downgrades require a brief reboot to apply the new resource limits, scheduled at a time you choose.
Moving between tiers (Shared CPU to Dedicated CPU, or back to Basic) involves a server move, which our team handles for you. There is no fee for upgrades or tier changes, you just pay the difference for the new plan.
30-day money-back guarantee on all Shared CPU plans. If you sign up and decide we are not the right fit within the first 30 days, we refund your full first payment, no questions asked, no exit fees, no contracts to cancel.
After 30 days, plans are billed monthly or annually depending on your selected cadence. You can cancel anytime through your account dashboard. Annual plans get pro-rated refunds for unused months on cancellation.