Rack and Stack Services in Brazil

Rack and stack refers to the physical process of installing servers, storage arrays, networking equipment, and supporting infrastructure inside datacenter racks. The term covers the complete sequence from equipment receipt at the loading dock through to a powered, cabled, and documented deployment ready for remote OS provisioning. A full rack and stack engagement at a Brazilian colocation facility typically involves seven steps: pre-staging and inventory verification, rail kit and mounting hardware preparation, physical server mounting in the correct rack unit positions, structured cable runs (Cat6a copper or OM4 fibre, depending on the interconnect design), power circuit connections to 220V/60Hz PDUs, initial power-on and BIOS-level health checks, and post-installation documentation covering rack diagrams, cable labels, and asset records. Reboot Monkey performs all seven steps on-site across Brazil's major datacenters. Our field engineers work inside third-party colocation facilities including Equinix SP1 in the Sao Paulo metro, Ascenty facilities in Sao Paulo and Campinas, and Odata sites in Sao Paulo and Rio de Janeiro. We do not own or operate these facilities. We work inside them on behalf of enterprises, financial institutions, and hyperscale tenants who need physical labour without the overhead of maintaining a resident installation team. For IT directors managing hardware deployments remotely, the alternative to using a specialist rack and stack provider is either flying in your own engineers or relying on the datacenter operator's internal team. Datacenter operators typically charge premium rates for out-of-scope work and are bound by their own scheduling queues. Reboot Monkey operates independently and can mobilise within a 4-hour response window for priority deployments at any supported Brazilian facility.

Brazil is the largest datacenter market in Latin America. The Sao Paulo metropolitan region hosts the highest concentration of carrier-neutral colocation capacity in South America, anchored by the IX.br internet exchange, which interconnects over 2,100 autonomous systems at peak throughput above 18 Tbps, according to NIC.br data (2025). This traffic density makes Sao Paulo the natural hub for enterprises requiring low-latency connectivity to Brazilian end users. The key operators in the Brazilian market are Equinix (Sao Paulo metro campus), Ascenty (majority-owned by Digital Realty, with facilities in Sao Paulo, Campinas, and Fortaleza), Odata (Sao Paulo and Rio de Janeiro), and Host.net Datacenter. Rio de Janeiro supports a secondary market driven by the oil-and-gas sector and state government workloads. Campinas is a growing alternative to Sao Paulo, offering lower land costs and access to the UNICAMP technology cluster. For rack and stack purposes, the Brazilian facility landscape creates specific operational requirements. Brazilian power infrastructure delivers 220V at 60Hz, which differs from the 230V/50Hz standard common in continental Europe. Field engineers must verify PDU specifications before connecting any equipment, and power strips purchased outside Brazil may require replacement if they cannot accommodate local voltage and frequency tolerances. Reboot Monkey engineers are familiar with these requirements and verify power compatibility as a standard pre-installation step. Hardware logistics are also a practical consideration for Brazilian deployments. Import duties on computing equipment can be significant under Brazil's current tariff structure, which sometimes leads enterprises to procure locally or use bonded warehouse arrangements. Our field engineers can work with bonded inventory received at facility loading docks and perform installation once equipment clears customs, reducing idle rack time.

The surge in AI infrastructure investment is reshaping rack and stack requirements across every major market, and Brazil is no exception. Brazilian financial institutions, technology companies, and cloud providers have accelerated GPU server deployments since 2024, creating demand for high-density rack installations that exceed the capability of standard enterprise rack and stack processes. High-density deployments at 30kW to 100kW per rack require a different approach than traditional 5-10kW server installations. The physical work itself is more demanding: an NVIDIA DGX H100 system, for example, weighs approximately 63kg and requires at minimum two trained engineers for safe rack insertion. Rail kit alignment is critical at this weight, and improper installation risks equipment damage and voided warranties. Reboot Monkey assigns two-person teams for any single unit exceeding 30kg as a standard safety protocol. Cabling at high density also differs substantially from standard copper-centric deployments. GPU clusters typically use Direct Attach Copper (DAC) cables for short rack-to-rack links at 400G or 800G speeds, with Active Optical Cables (AOC) for longer runs between racks or to top-of-rack switches. Each cable type has specific bend radius and minimum distance requirements. Our engineers follow structured cabling standards throughout, labelling every cable at both ends, maintaining consistent colour coding across the deployment, and documenting cable routes in the post-installation record. Power distribution for GPU racks in Brazilian facilities must account for the 220V/60Hz supply characteristics alongside the high current draw of modern accelerators. Our field teams confirm breaker sizing and PDU ratings before installation and coordinate with the facility operations team if power circuit upgrades are required ahead of the deployment window. For enterprises deploying AI infrastructure in Brazil, <a href="/en/smart-hands/brazil/">smart hands support</a> is often required alongside the initial rack and stack to handle firmware updates, network configuration, and initial system validation after power-on. Reboot Monkey provides both services under a single engagement, removing the coordination overhead of managing separate vendors for physical and technical tasks.

Brazil's Lei Geral de Proteção de Dados (LGPD) governs personal data processing and has direct implications for enterprises operating IT infrastructure in Brazilian datacenters. The LGPD, enforced by the Autoridade Nacional de Proteção de Dados (ANPD), imposes fines of up to BRL 50 million per infraction under Article 52. In the event of a data incident, the LGPD requires notification to the ANPD within 3 business days of the controller becoming aware of the incident, per ANPD Resolution CD/ANPD No. 2/2022. For rack and stack operations specifically, LGPD compliance intersects primarily with asset documentation and decommissioning practices. When Reboot Monkey installs equipment that will process personal data, we provide a complete post-installation asset record: serial numbers, rack unit positions, power circuit assignments, network port connections, and cable labels. This documentation supports the data controller's ability to demonstrate which physical assets hold personal data, which is a practical requirement when responding to subject access requests or data erasure obligations. The Conselho Monetário Nacional Resolution 4.658/2018 (BACEN CMN 4.658) governs cloud and data processing arrangements for Brazilian financial institutions. This regulation addresses how financial institutions contract for data processing and cloud services. It does not restrict physical access to datacenters by authorised third-party technicians. Rack and stack activities performed by Reboot Monkey on behalf of a financial institution tenant are physical infrastructure services that fall outside the regulatory scope of CMN 4.658/2018, which targets data processing and cloud service contracts rather than on-site hardware installation. Post-installation documentation is a standard deliverable for every Reboot Monkey rack and stack engagement in Brazil. Clients receive a rack diagram, a cable schedule, an asset register with serial numbers and rack unit positions, and confirmation that all equipment powered on successfully. This documentation supports audit readiness under both LGPD and any sector-specific compliance frameworks the client operates under. <a href="/en/contact/">Contact Reboot Monkey</a> to discuss documentation requirements for your specific compliance context before your installation date.

Structured cabling is the backbone of any reliable datacenter deployment, and the quality of the cabling work performed during rack and stack directly determines how easy or difficult the infrastructure will be to maintain, expand, and troubleshoot over its operational lifetime. Reboot Monkey field engineers follow a structured cabling methodology derived from TIA-942 datacenter cabling standards. For copper horizontal runs, we use Cat6a or better to support 10GbE to the server and 100GbE at aggregation layers. For inter-rack fibre, OM4 multimode supports 100GbE up to 150 metres and 400GbE up to 100 metres, making it suitable for the majority of Brazilian colocation deployments. Where clients specify single-mode fibre for longer campus runs or future-proofing to 400G-ZR transceivers, we accommodate OS2 single-mode installations. Every cable is labelled at both termination points with a consistent naming convention agreed with the client before installation begins. Labels include the rack identifier, patch panel port, and destination device. Cable management arms are fitted to all rack-mount servers where the chassis supports them, and horizontal cable managers are installed between patch panels and switch ports to maintain bend radius and allow individual cable removal without disturbing adjacent runs. Post-installation documentation is delivered within 24 hours of installation completion and includes: - Rack elevation diagram (front and rear views, to scale, with rack unit assignments) - Cable schedule listing every cable by label, origin port, destination port, cable type, and length - Power circuit map listing each device, its PDU inlet, breaker number, and measured draw at initial power-on - Asset register with manufacturer, model, serial number, rack unit position, and installation date for every installed component - Power-on verification log confirming BIOS access and hardware health status for each server This documentation package is delivered electronically in PDF and Excel formats. Clients with CMDB integrations can receive data in CSV or JSON formats to enable direct import into ServiceNow, Device42, or equivalent asset management platforms. Reboot Monkey also provides <a href="/en/remote-hands/brazil/">remote hands support</a> for any follow-up cabling corrections or additions identified during the first operational weeks after deployment.

Rack and stack and smart hands are related but distinct services, and understanding the difference helps you scope the right engagement before you contact a vendor. Rack and stack is fundamentally a physical installation service. It covers the mechanical and electrical work of installing hardware into racks: mounting, cabling, power connection, and initial power-on verification. The scope ends at the point where hardware is physically installed and confirmed operational at the BIOS level. Network configuration, operating system installation, application deployment, and ongoing monitoring are outside the rack and stack scope. Smart hands is a broader, more technical service covering configuration, troubleshooting, and skilled technical tasks that require IT expertise beyond physical installation. A smart hands technician can follow a runbook to configure a switch, reload an operating system, validate a firewall policy, or replace a failed component following a specific procedure. Smart hands is the right service for tasks that require judgment and technical knowledge rather than physical labour. <table> <thead> <tr> <th>Capability</th> <th>Rack and Stack</th> <th>Smart Hands</th> </tr> </thead> <tbody> <tr> <td>Physical hardware mounting</td> <td>Yes</td> <td>Yes</td> </tr> <tr> <td>Structured cabling</td> <td>Yes (full scope)</td> <td>Yes (patching only)</td> </tr> <tr> <td>Power circuit connection</td> <td>Yes</td> <td>Yes</td> </tr> <tr> <td>BIOS-level power-on check</td> <td>Yes</td> <td>Yes</td> </tr> <tr> <td>Network configuration</td> <td>No</td> <td>Yes (runbook-guided)</td> </tr> <tr> <td>OS installation</td> <td>No</td> <td>Yes (runbook-guided)</td> </tr> <tr> <td>Break-fix and component swap</td> <td>No</td> <td>Yes</td> </tr> <tr> <td>Post-install documentation</td> <td>Yes (full package)</td> <td>Yes (task-specific)</td> </tr> <tr> <td>Typical response SLA</td> <td>4 hours</td> <td>4 hours</td> </tr> </tbody> </table> For a greenfield hardware deployment, start with rack and stack. If your deployment also requires OS provisioning or network configuration, add <a href="/en/smart-hands/brazil/">smart hands support</a> to the same engagement. Reboot Monkey coordinates both services under a single project manager so you deal with one point of contact rather than separate service orders. For enterprises planning a larger infrastructure move, <a href="/en/server-migration/brazil/">server migration services</a> cover the full lifecycle including decommissioning at the source facility, physical transport, and rack and stack at the destination.

Brazil's rack and stack market is shaped by several dominant industry verticals, each with distinct requirements. Financial services is the largest driver of colocation demand in Brazil. The country's banking sector operates one of the most advanced real-time payments systems in the world (PIX, launched by the Banco Central do Brasil in 2020 and processing over 4 billion transactions monthly as of 2025, according to BCB statistics). The transaction volumes and latency requirements of financial infrastructure demand on-premises or colocation deployments in Sao Paulo rather than reliance on public cloud alone. Financial institutions regularly expand hardware capacity in colocation facilities, generating recurring rack and stack demand. The technology and e-commerce sector, led by companies in the Sao Paulo metro, drives significant hardware deployment activity aligned with growth cycles and seasonal peaks. Streaming and media companies serving the Brazilian and broader Latin American market also maintain significant colocation footprints, particularly at IX.br-connected facilities where the network density supports content distribution at scale. The oil and gas sector, centred in Rio de Janeiro, operates substantial on-premises and colocation infrastructure for seismic data processing and operational technology systems. These deployments often involve specialised hardware with high power and cooling requirements, making an experienced rack and stack team essential. Healthcare is an emerging growth vertical as Brazil's health technology sector digitises records and centralises data processing. The Lei Geral de Proteção de Dados applies to health data as a special category, making documentation practices in rack and stack engagements particularly important for healthcare clients. Contact Reboot Monkey at <a href="/en/contact/">our contact page</a> to discuss requirements specific to your industry before scheduling your installation. For enterprises planning a larger infrastructure lifecycle, including retirement of existing hardware, <a href="/en/data-center-decommissioning/brazil/">datacenter decommissioning services</a> in Brazil cover secure asset removal, ITAD coordination, and end-of-life documentation.