Latitude.sh c2.small.x86, $ 92.00/mo. on Linux Dedicated

c2.small.x86 has been added on Jun 16, 2025

💪 CPU/Cores :	Intel Xeon E-2234 4C/8T @3.6GHz
🔋 RAM :	32000 MB
🔌 Hosted domains :	unlimited
🆓 free domains :	0
📌 Dedicated IPs :	0
💳 Payment Methods :	Credit / Debit / Prepaid CardsPayPal
🔧 Category :	Self Managed
✍️ Support Options :	Email
🌏 Server Locations :	Argentina Australia Brazil Germany Japan Mexico Netherlands Singapore United Kingdom United States
⚑ Targeting :	BR US

See also initial c2.small.x86 plan location on their website!

*📜 Plan description

📄 Editorial Review

Latitude.sh is a global infrastructure provider focused on automated bare-metal servers, GPU computing, virtual machines, managed Kubernetes, storage, managed PostgreSQL, and private network connectivity. The business began in 2001, when brothers Guilherme Soubihe and Eduardo Soubihe founded Maxihost in Brazil to supply website and email hosting services to small and medium-sized companies.

The company gradually moved beyond traditional web hosting as its customers required more powerful and flexible infrastructure. Maxihost became Latitude.sh on August 16, 2022, adopting a name more closely associated with Linux, shell environments, automation, and globally distributed computing. Guilherme Soubihe remained CEO, while the company retained its Brazilian roots and expanded across North America, Europe, Latin America, Asia, and Australia.

Latitude.sh became part of Megaport (megaport.com) on November 27, 2025, following an acquisition agreement announced earlier that month. At the time, the company operated more than 7,700 servers, served over 1,150 customers, and maintained infrastructure in 20 major markets across 10 countries. Latitude.sh continues to operate under its established brand, while its integration with Megaport expands the private-network connectivity available to customers.

This is not a conventional shared-hosting company. Latitude.sh does not concentrate on cPanel hosting packages, reseller accounts, domain registration, website builders, or managed WordPress services. Its infrastructure is better suited to developers, SaaS companies, artificial-intelligence businesses, infrastructure engineers, gaming platforms, blockchain projects, streaming providers, ad-tech companies, and enterprises requiring predictable CPU, GPU, storage, and network performance.

Metal is the central bare-metal service. Each customer receives a physically isolated, single-tenant server instead of a virtualized portion of shared hardware. Configurations include AMD EPYC and Intel processors, substantial memory allocations, local NVMe storage, and network interfaces ranging from 10 Gbps to dual 100 Gbps. In-stock systems can usually be provisioned within seconds through the dashboard or API.

The operating-system selection includes Ubuntu, Debian, Rocky Linux, Red Hat Enterprise Linux, Windows Server, and other supported images. Server-management options include RAID 0 and RAID 1, SSH-key deployment, operating-system reinstallation, rescue mode, serial-console access over SSH, remote IPMI management, and custom operating-system deployment through iPXE.

Metal GPU provides dedicated NVIDIA GPU servers for artificial intelligence, machine learning, inference, model training, rendering, and other accelerator-intensive workloads. Available systems range from machines with a single H100 GPU to large configurations containing eight RTX PRO 6000 GPUs. GPU resources remain dedicated to one customer rather than being divided through a shared or spot-instance model, giving demanding workloads more consistent performance.

Virtual Machines provide a smaller and more affordable entry point into the platform. They combine virtualized processors and memory with local storage and a 10 Gbps network interface. Available sizes start with four vCPUs and 16 GB RAM and scale to 16 vCPUs and 64 GB RAM. Billing continues while a VM is powered off, so the resource must be deleted to stop further charges. Resizing is currently limited to upgrades, making initial capacity planning important.

Latitude Kubernetes Service, also known as LKS, creates RKE2 Kubernetes clusters on dedicated bare-metal infrastructure. Control-plane and worker nodes can be deployed through the dashboard or API, while MetalLB supplies BGP-based LoadBalancer addresses. A cluster can use between one and three control-plane nodes and as many as ten worker nodes.

LKS remains a preview service. Customers are responsible for day-two operations such as ingress configuration, persistent storage, monitoring, application security, backups, and operating-system maintenance. It is therefore better suited to teams already familiar with Kubernetes administration than to customers looking for a completely managed container platform.

Build is the enterprise service for customized infrastructure. Customers can request specific processors, memory capacities, storage layouts, network switches, spare-hardware policies, server quantities, and deployment regions. Projects can range from several machines to thousands of systems. New custom locations may be prepared in approximately 45 days when logistics, hardware availability, and facility capacity allow it.

Enterprise deployments can include dedicated account management, custom service-level agreements, shared Slack communication, and replacement-hardware arrangements. This makes Build particularly relevant for companies that need standardized infrastructure across several regions or require configurations that are not available through the regular server catalogue.

Storage includes object, block, and file-storage services. Object storage uses an S3-compatible API and is suitable for AI datasets, backups, archives, analytics, media libraries, and other unstructured data. Standard and high-performance object-storage tiers are available.

Block storage supplies persistent volumes that remain separate from the lifecycle of a virtual machine. File storage provides shared NFS file systems that can be accessed by multiple compute instances. Availability differs by location, and some block or file-storage functions remain subject to early-access approval.

Managed PostgreSQL removes much of the operational work normally associated with database administration. It includes automatic backups, point-in-time recovery, software patching, version upgrades, monitoring, alerts, multi-availability-zone architecture, and automatic failover. This is a practical option for customers that need reliable PostgreSQL infrastructure without maintaining replication, backup, and recovery systems themselves.

Cloud Gateway creates private Layer 3 connections between Latitude.sh infrastructure and AWS, Microsoft Azure, or Google Cloud. It supports hybrid and multi-cloud architectures without forcing all communication to cross the public internet.

Global Gateway connects Latitude.sh regions through a private backbone with intercontinental capacity reaching 400 Gbps. Shared backbone connectivity is available for ordinary deployments, while dedicated capacity can be arranged for larger projects. Private RFC1918 addressing allows servers in different regions to communicate without exposing internal services through public IP addresses.

Additional networking functions include same-location Layer 2 private VLANs, Elastic IP addresses, floating IP addresses, BGP, BYOIP, custom ASNs, IPv4 and IPv6 allocations, centralized firewall rules, and free private traffic. DDoS protection is included with servers and covers Layer 3, Layer 4, and Layer 7 attacks through a distributed scrubbing platform with as much as 7 Tbps of mitigation capacity. Exceptionally large attacks may still result in temporary null-routing when continued mitigation could affect the wider network.

The website and management environment are modern, responsive, and strongly oriented toward technical users. The main website and documentation use English, while product prices are primarily displayed in US dollars.