Linode (Akamai)

What Linode (Akamai) actually is

Linode is the developer cloud that made cheap Linux VMs respectable, and since March 2022 it has been part of Akamai, which bought the company for about $900 million and now sells it as Akamai Cloud. The product set is familiar: Linux virtual machines (Linodes), block and object storage, managed databases, Kubernetes (LKE), NodeBalancers, DNS, and — newer since the acquisition — GPU instances and an inference line. The interface, API, and CLI are still recognizably Linode, and the pricing is still published as a table rather than a calculator you have to model.

The core offering is the compute instance lineup. Shared CPU plans run from the Nanode 1 GB at $5.00 per month ($0.0075 per hour) with 1 vCPU, 1 GB RAM, 25 GB SSD, and 1 TB of transfer, up through the classic ladder — Linode 2 GB at $12, 4 GB at $24, 8 GB at $48, 16 GB at $96, 32 GB at $192, and larger shapes at $384 and beyond — with per-hour billing capped monthly. Dedicated CPU plans start at $36 per month for 4 GB on G6 hardware ($43 on G7), High Memory at $60 per month for 24 GB, and the GPU line covers NVIDIA RTX 4000 Ada at $350 per month, RTX 6000 Quadro at $1,000, and a single-card RTX PRO 6000 Blackwell at $1,665. Transfer allowances are pooled across the account, and overage beyond the pool is $0.005 per GB — one of the lower egress rates among the US clouds.

What the Akamai ownership actually changed is underneath. Instances now sit on Akamai's network, the company is pushing workloads toward its edge footprint, and security and delivery products from the parent catalog are one integration away. What it did not change is the shape of the product: this is still a Linux-VM shop with an API-first workflow, not a hyperscaler with 200 services.

Pricing: what you really pay

The published numbers are honest and complete enough to plan with, which is why Linode has stayed on our shortlists. Billing is hourly with a monthly cap — the pricing page itself walks through the example of a 1 GB VM deleted after 24 hours costing $0.18 — and invoices land on the first of the following month. Region does not change compute list pricing the way it does at some competitors, though object storage availability does vary by region.

Where the bill grows is the services around the VM. Backups are a paid add-on starting at $2 per month for a 1 GB Shared CPU instance (roughly 20% of instance price at larger sizes). Managed databases start around $15 per month, NodeBalancers around $10, and LKE clusters bill for their worker nodes. A $5 minimum fee applies to accounts with under 250 GB of storage. Linode Managed, the hands-on operations service, is $100 per Linode per month and includes the Backups service at no extra charge.

We can put real numbers on the entry plan. On 1 April 2026 we tested a Nanode 1 GB in Frankfurt — 1 vCPU, 1 GB RAM, 25 GB SSD, at the €5/month list price — with sysbench, fio, and iperf3: a CPU score of 1,800, about 14,000 IOPS, and roughly 800 MB/s of network throughput. For context in our benchmarks, that places the Nanode where its price suggests: adequate for small services and development, but clearly behind the faster NVMe-backed entry plans that European specialists sell at similar prices. You buy the platform, the API, and the SLA with this $5; the raw silicon is not the bargain.

Performance & infrastructure

Akamai Cloud lists 25-plus core compute data centers: ten in North America (Atlanta, Chicago, Dallas, Fremont, Los Angeles, Miami, Newark, Seattle, Toronto, Washington DC), seven in Europe (Amsterdam, Frankfurt, London, Madrid, Milan, Paris, Stockholm), seven in Asia-Pacific (Chennai, Mumbai, Osaka, Singapore, Tokyo, Melbourne, Sydney), plus São Paulo and Jakarta. Distributed compute locations extend the edge story further, but note the product split: some services, including Backups, exist only in core regions. If you plan around a distributed location, check service availability before you design for it.

The reliability story is contractual and unusually legible. The compute SLA (current version 1.2, effective 27 February 2026) guarantees 99.99% monthly uptime for covered compute services in general availability — 99% for instances in limited availability — with pro-rata service credits claimed through support within 30 days. Everything else — Backups, Block Storage, Object Storage, Kubernetes, DNS, NodeBalancers, Managed Databases — is covered at 99.9%. Exclusions are the usual ones (scheduled maintenance with 24 hours' notice, force majeure, DDoS events, customer software), but the two-tier structure is more honest than a single marketing number.

Support runs 24/7 by ticket, phone, and chat, with the service-credit mechanism giving the SLA teeth. Long-time customers will tell you support quality has fluctuated through the Akamai integration, and the community forums are active enough that most configuration questions are answered publicly. The documentation — techdocs.akamai.com, still branded Linode in places — remains among the best in the industry: task-focused, current, and honest about limitations (the Backups page, for instance, plainly lists its own restrictions).

Ecosystem-wise, there is an app marketplace for one-click deployments, a well-documented REST API, an official CLI, Terraform support, and managed offerings (LKE, Managed Databases, Linode Managed) for teams that want the platform to run more of the stack. For a small team standardizing on API-driven infrastructure, this is a complete toolbox.

Where it falls short

Raw price-performance is the recurring weakness. Our Nanode measurements show an entry instance that is merely fine: the shared plans are priced above the European value leaders, and the older shared hardware behind the low tiers is explicitly described as legacy in the pricing pages. Teams chasing the best CPU or IOPS per euro will find better elsewhere, and the same pattern holds one tier up — dedicated plans are competent but not aggressive.

The second issue is the diffusion of the product. Since the Akamai acquisition, the platform has absorbed GPU lines, inference services, and edge positioning while some older Linode conveniences have aged. Customers report account and billing transitions, shifting product names (Akamai Cloud versus Linode), and a sense that the roadmap serves Akamai's enterprise strategy first. Nothing here is broken, but the developer-first identity is quieter than it was.

Service availability is uneven across regions: Backups are core-regions-only, object storage availability varies, and distributed compute locations carry different capability sets. That matters for anyone porting an architecture between regions.

Finally, several resilience features are add-ons rather than inclusions. Backups cost extra, snapshots come with the Backups service rather than for free, and dedicated DDoS protection is a separate Akamai product line rather than a feature stamped on every instance. A $5 VM plus backups plus a managed database is a very different bill from the headline.

Who should pick this

We would pick Akamai Cloud (Linode) for teams that want a mainstream, API-first Linux platform with a real SLA, broad regional coverage, and a parent company that owns one of the largest networks on earth: production web services with pooled transfer, Kubernetes estates on LKE, managed databases without self-hosting, and workloads that may eventually want edge delivery or CDN services from the same vendor. The 99.99% compute SLA with service credits and the per-hour billing make it an easy recommendation for small production teams.

We would not pick it as the budget default. If compute per euro is the whole story, the entry plans lose to the price-performance leaders in our vps-providers category, and our own benchmark numbers on the Nanode back that up. Nor is it the right home for teams wary of a large parent company's strategic pivots. For everything between those poles, Linode remains one of the safer middle choices — predictable, documented, and contractually backed.

Key takeaways

• Akamai-owned since March 2022 ($900M); the developer-friendly Linode product survives inside Akamai Cloud
• Published hourly pricing with monthly caps: Nanode 1 GB at $5/mo up to GPU plans at $1,665/mo; egress overage $0.005/GB
• We tested a Nanode 1 GB in Frankfurt on 2026-04-01: CPU 1,800, 14,000 IOPS, 800 MB/s at €5/mo — adequate, not class-leading
• 99.99% compute uptime SLA with pro-rata service credits; 99.9% for storage, Kubernetes, databases, and other services
• 25+ core data centers across four continents; some services (e.g. Backups) only in core regions
• Resilience costs extra: Backups from $2/mo, managed databases from ~$15/mo, DDoS protection is a separate Akamai product

Sources

• https://www.akamai.com/cloud/pricing
• https://www.akamai.com/legal/sla-compute
• https://www.akamai.com/legal/sla-non-compute
• https://techdocs.akamai.com/cloud-computing/docs/backup-service
• https://linode.com/docs/products/services/managed/guides
• http://en.wikipedia.org/wiki/Linode
• https://cloudpicked.com/linode-akamai-review-en.html
• https://costbench.com/software/cdn-edge/akamai
• https://www.linode.com/press-release/akamai-to-acquire-linode
• https://vpsbenchmarks.com/hosters/linode
• https://techdocs.akamai.com/cloud-computing/docs/an-overview-of-ipv6-on-linode