Linode (Akamai), now operating as Akamai Cloud Computing following the acquisition, prices its Shared CPU line on a flat-rate ladder that steps up in compute, RAM, and storage at each named plan, from its smallest Nanode to its largest Shared CPU instance, all billed hourly with a monthly cap. The sticker price on that ladder is simple to read. What is not on the plan card is what happens once a workload crosses its included transfer, or once a team attaches a load balancer, a managed Kubernetes control plane, or Linode's managed-service layer, none of which show up next to the instance price. This page works through the ladder itself, then the parts of the bill that live outside it. That distinction matters more on Linode than on some competitors, because the ladder pricing itself is unusually predictable: there is little ambiguity about what a given tier costs per month. The estimate only goes wrong once egress, load balancing, or managed add-ons enter the picture, which is exactly the territory the rest of this page covers.
This page covers Linode's Shared CPU compute line only: the Nanode and Linode-branded plans priced in the ladder below. Akamai also sells Dedicated CPU, GPU, and bare-metal tiers under the same Cloud Computing umbrella; those carry separate rate cards that this page's facts do not cover, so treat every figure here as Shared CPU pricing, not a whole-catalog number. This is also a single-entity page. It does not compare Linode against DigitalOcean, Vultr, or any other provider in this comparison; that cross-provider math lives on the real-cost flagship. Every price below is the on-demand, hourly-capped rate; committed-use or reserved pricing, if Linode offers it, is out of scope for this page.
The Shared CPU pricing ladder, tier by tier
Every tier below runs the same underlying Shared CPU instance type; only the resource allocation and price change. Hourly billing caps at the monthly figure shown, so a Linode left running for a full month never costs more than the number in the Monthly column no matter how many hours it accrues.
| Plan | Monthly Price | vCPUs | RAM | Storage | Transfer Included | HD Verdict |
|---|---|---|---|---|---|---|
| Nanode (entry tier) | $5.00 | 1 CPUs | 1 GB | 25 GB | 1 TB | Entry tier for a single low-traffic site or dev box. |
| Linode, 2nd tier | $12.00 | 1 CPUs | 2 GB | 50 GB | 2 TB | Small production WordPress site or API. |
| Linode, 3rd tier | $24.00 | 2 CPUs | 4 GB | 80 GB | 4 TB | Small business site with headroom for a staging copy. |
| Linode, 4th tier | $48.00 | 4 CPUs | 8 GB | 160 GB | 5 TB | Multi-site hosting or a light application server. |
| Linode, 5th tier | $96.00 | 6 CPUs | 16 GB | 320 GB | 8 TB | Database or application tier under real load. |
| Linode, 6th tier | $192.00 | 8 CPUs | 32 GB | 640 GB | 16 TB | Last tier where included transfer still scales with price (see below). |
| Linode, 7th tier | $384.00 | 16 CPUs | 64 GB | 1280 GB | 20 TB | Transfer allowance plateaus from here up the ladder. |
| Linode, 8th tier | $576.00 | 20 CPUs | 96 GB | 1920 GB | 20 TB | Compute-heavy workload, not a transfer-heavy one. |
| Linode, 9th tier | $768.00 | 24 CPUs | 128 GB | 2560 GB | 20 TB | Same transfer cap as the tiers below and above it. |
| Linode, flagship tier | $1,152.00 | 32 CPUs | 192 GB | 3840 GB | 20 TB | Flagship Shared CPU tier; weigh Dedicated CPU at this budget too. |
What the included transfer actually buys as you climb the ladder
Linode's transfer allowance is where the ladder earns its reputation as generous, though not evenly across all ten tiers. From the entry Nanode tier through Linode's 6th tier, the included transfer climbs alongside the monthly rate at every step: 1 TB on the entry tier, 4 TB at the 3rd tier, 16 TB once you reach the 6th tier. From the 7th tier up, that pattern breaks. The transfer allowance holds flat at 20 TB across the 7th through 9th tiers, and it is still 20 TB on the flagship tier (the identical figure), even though the monthly rate keeps rising from $384.00 up to $1,152.00 across those same four tiers.
In practice, that means transfer-per-dollar is best at the bottom of the ladder, not the middle: the entry Nanode tier returns more included transfer for every dollar spent than any tier above it, and that ratio erodes at every step up from there. What actually holds through the 6th tier is a different, narrower property: included transfer keeps growing in step with price all the way to that point, so a transfer-heavy workload still gets proportional transfer growth if it scales no higher than the 6th tier. Past that point the pattern breaks: the transfer allowance plateaus (see the 7th through 9th tiers above) while the price keeps climbing, so a workload that is transfer-heavy but not especially compute-heavy is better served running several 6th-tier-or-lower Linodes side by side than scaling vertically into the top three tiers, where the extra money buys CPU and RAM headroom the transfer allowance no longer grows to match. Budget the top of the ladder for compute-bound work, not bandwidth-bound work.
Two overage rates, not one
Once a workload exceeds its plan's included transfer, Linode bills a different rate depending on where the traffic actually leaves the network, and averaging the two together is the easiest way to under-budget a Linode invoice. Compute egress, meaning traffic leaving a Shared CPU, Dedicated CPU, or GPU instance directly, is billed at US$0.005 per GB. Traffic leaving through a NodeBalancer (Linode's managed load balancer, priced separately below) is billed at US$0.015 per GB instead, a distinctly higher per-gigabyte figure for what is, from a buyer's seat, the same category of outbound traffic.
| Traffic Source | Overage Rate | HD Verdict |
|---|---|---|
| Compute instance egress (Shared CPU / Dedicated CPU / GPU) | US$0.005 per GB | Applies to a single Linode serving traffic directly. |
| NodeBalancer egress | US$0.015 per GB | Applies once traffic routes through a load balancer; budget it separately, never average the two rates together. |
The gap matters most for anyone running a load-balanced setup: multiple Linodes behind a NodeBalancer for redundancy or horizontal scaling, a common pattern once a single instance stops being enough for a production workload. Every gigabyte that crosses the NodeBalancer instead of leaving an instance directly is billed at the higher of the two rates. Architecting for high availability with a NodeBalancer in front of the instances means modeling overage at the NodeBalancer rate, not the compute rate. Treating them as one number understates the bill for exactly the workloads sophisticated enough to need a load balancer in the first place.
On the invoice itself, compute overage and NodeBalancer overage post as separate charges tied to separate resources, not a single blended bandwidth line, so a team reconciling a monthly Linode bill against its own cost model should expect to see, and budget for, both rates independently rather than hunting for one combined "overage" figure to explain the gap from the ladder price.
The line items that don't show up on the instance price card
Beyond the Shared CPU ladder and its overage rates, several add-on products carry their own flat monthly charges once production infrastructure actually needs them:
- NodeBalancer (Linode's managed load balancer): $10 per month, plus its own egress overage rate covered above.
- LKE managed Kubernetes, HA control plane: US$60 per cluster, per month, charged on top of the worker-node instance pricing from the ladder above.
- LKE Enterprise control plane: US$300 per cluster, per month, the step up from HA for teams running Kubernetes at a larger, more regulated scale.
- Managed Service (Akamai's infrastructure-management add-on): $100 per compute instance, per month, charged per compute instance, a recurring line item that scales with fleet size rather than a one-time setup fee.
None of these are optional once a team needs the capability behind them. A production Kubernetes cluster with a highly-available control plane, for instance, isn't a single ladder-tier decision at all: it's the worker-node tier plus the control-plane charge, every month, for as long as the cluster runs. Stacking two or three of these add-ons against a mid-tier ladder plan is common for any team running real production infrastructure rather than a single hobby instance. A load-balanced pair of Linodes behind a NodeBalancer with a managed Kubernetes cluster in front adds the NodeBalancer charge and the control-plane charge on top of every instance's ladder price, before either overage rate is even considered.
Who the Linode ladder actually fits
The bottom of the ladder, from the entry Nanode tier through the 3rd tier, suits a single low-to-moderate-traffic site or a development box, where the included transfer at those tiers comfortably covers ordinary usage without touching either overage rate. The middle of the ladder, the 4th tier through the 6th tier, is the last stretch where included transfer still grows in step with price before the allowance plateaus, and it is the range that fits a small business running one production application with room to grow. Past the 7th tier, the calculus shifts: the extra money buys compute and RAM headroom, not more transfer, so that end of the ladder only makes sense for a workload that is actually CPU- or memory-bound. And any team running a load-balanced, multi-instance setup needs to price the NodeBalancer and its overage rate as a real, recurring part of the bill, not an afterthought bolted on after the instance total. Teams evaluating Linode purely on the ladder's entry price, without accounting for these add-ons, risk comparing an incomplete number against a competitor's fuller one. The ladder price is a real starting point, not the full monthly total for anything beyond the simplest single-instance setup.
Where to go next
For the cross-provider version of this math, comparing Linode's real cost against DigitalOcean, Vultr, and the rest of this comparison's certified pricing, see The Real Cost of Cloud VPS Hosting. To see where Linode fits among the full set of providers this cluster covers, start at the Cloud VPS Hosting hub.