RedCLARA built a continental backbone through federation, not ownership

Summary

• RedCLARA is a Uruguay-based nonprofit that federates national research and education networks; it is neither a shareholder-owned company nor a carrier controlling every campus connection.
• Eight principal routed nodes and long-term international capacity form a 100G-class backbone, with reported 100, 200 and 300 Gbps links and a smaller Central American segment.
• ALICE and ALICE2 created the institution and first network; BELLA secured long-term capacity on the direct Portugal–Brazil route, reducing dependence on indirect paths through North America.
• BELLA II’s test is converting project finance into durable member-supported infrastructure while strengthening weaker national networks, improving evidence and preventing supplier or membership bottlenecks.

BELLA changed the geography of the network

BELLA, the Building the Europe Link to Latin America programme, combined a submarine component with terrestrial upgrades. BELLA-S secured long-term optical spectrum on a direct cable between Portugal and Brazil. BELLA-T strengthened the South American network required to carry that capacity from the landing environment through regional nodes and national networks.

The physical subsea system became EllaLink, an approximately 6,000-kilometre direct route between Sines in Portugal and Fortaleza in Brazil. Production research-and-education traffic began in 2021. The programme initially provisioned two 100 Gbps links: one connecting GÉANT and RedCLARA and another supporting Copernicus Earth-observation connectivity. The broader infrastructure described in RedCLARA’s 2024 report extends far beyond the subsea span and is reported at 35,830 kilometres with Nx100G capability.

These lengths describe different things. Six thousand kilometres refers to the direct submarine span. The larger figure encompasses terrestrial and wider programme infrastructure. Likewise, project costs differ because official documents refer to different legal actions and programme boundaries. CORDIS records €12.4447 million for the BELLA-S1 action, including €10 million in EU funding, while wider programme accounts include terrestrial work, multiple European funding lines and national contributions. These numbers should not be collapsed into one unexplained total.

BELLA’s practical value was route choice and upgradeable capacity. Programme sources have reported latency reductions of up to 50% for some Europe–Latin America paths. That claim should remain path specific and attributed; it is not a universal performance result. The durable achievement is more fundamental: the research community gained a direct South Atlantic route and long-term capacity rights that could be expanded as optical technology and demand evolved.

Forty optical channels do not mean owning the cable

BELLA obtained long-term rights to 40 optical channels on the EllaLink system. This is one of the most important facts in RedCLARA’s infrastructure story and one of the easiest to overstate. A spectrum entitlement can provide substantial control over how capacity is lit and upgraded, but it does not make RedCLARA or GÉANT the sole owner of the commercial cable, the landing stations or every supporting facility.

A simplified chain runs from fibre pair to spectrum entitlement, from spectrum to provisioned optical channels, from those channels to 100G or later coherent transmission, and from the landing station into terrestrial backbones. Each layer has separate responsibilities. The cable operator maintains the marine system; landing and colocation providers support facilities; RedCLARA and GÉANT coordinate research capacity; equipment vendors supply optical systems; national networks deliver traffic to institutions.

Long-term spectrum rights can be more valuable than a fixed managed circuit because capacity can be increased by changing transmission equipment within optical and contractual limits. They also expose the consortium to shared-cable governance, repair schedules, power, spectrum planning and equipment interoperability. The right is durable, not self-sufficient.

This ownership distinction is central to an evidence-backed profile. RedCLARA can legitimately describe a regional backbone and a direct transatlantic route because it coordinates and operates capacity across them. It should not be described as owning every line on the network map. The organisation’s power lies in federation and contractual control of strategic capacity, not in complete physical vertical integration.

BELLA and the direct South Atlantic route

The economic change was equally important. Spectrum rights are not the same as a fixed managed circuit. They allow the partners to light optical channels and upgrade transmission equipment within contractual and physical limits. That creates a longer asset horizon and can separate capacity growth from repeated purchases of short-term bandwidth. It also creates continuing responsibilities for landing-station equipment, optical planning, maintenance, repair and technology refresh.

BELLA was not one grant with one universally valid cost. The specific BELLA-S1 action recorded by the European Commission had a total cost of approximately €12.44 million and an EU contribution of €10 million. Wider programme sources use other totals because they include terrestrial work, different European funding lines, national infrastructure and in-kind contributions. The correct lesson is not to select the largest number. It is that a regional cable and backbone programme is assembled from several legal and financial layers.

The direct route made RedCLARA more strategically important, but it did not make the organisation owner of the Atlantic. EllaLink remains a commercial cable system with its own stakeholders. RedCLARA and GÉANT hold long-term capacity and spectrum rights within that system. Their control is substantial at the service layer and limited at the underlying asset layer. That distinction is the foundation of an accurate infrastructure profile.

Spectrum rights, terrestrial reach and the ownership boundary

The phrase “RedCLARA network” can conceal several types of control. The organisation may own routers or other equipment at a point of presence. It may lease rack space and power. A national NREN may contribute fibre. A carrier may provide a managed segment. BELLA may supply a wavelength, an IRU or spectrum entitlement. A partner network may carry traffic beyond the regional boundary. The map combines these relationships into one usable service without converting them into one property title.

Spectrum rights on EllaLink are valuable because they provide a defined portion of the cable’s optical resource over a long period. The holder can provision channels as equipment and demand evolve, subject to the cable’s optical plan, coherent technology, landing-station systems and contract. The arrangement gives more control than buying a small circuit month by month, but less control than owning and operating the entire submarine system.

BELLA-T addressed the terrestrial problem. Landing a cable in Fortaleza does not connect a laboratory in Chile, Ecuador or Colombia. Traffic still has to move from the landing station through a regional point of presence, across the continental backbone, into a national NREN and finally to the institution. BELLA-T strengthened those South American paths and distributed the value of the direct Atlantic route beyond Brazil.

This chain explains why national networks remain indispensable. RedCLARA can coordinate the continental middle, but RNP, REUNA, RENATA, CEDIA and other NRENs supply the national reach and local relationships. An institution’s experience depends on the weakest link from campus to international destination. A powerful transatlantic segment cannot compensate for an underfunded national backbone or an inadequate campus connection.

Ownership language also affects risk analysis. A service based on an IRU may have different renewal and failure exposure from one based on a managed carrier circuit. A partner-contributed fibre segment may depend on the partner’s budget and civil works. A subsea spectrum right still depends on marine maintenance, landing stations and cable governance. Understanding these rights is more useful than asking whether RedCLARA “owns the network” in a simple yes-or-no sense.

The organisation’s most important asset may therefore be institutional rather than physical: the ability to assemble many rights, suppliers and national contributions into a coherent regional service. That ability creates bargaining power and continuity, but it also means contract management and partner alignment are part of the technical architecture.

A continental network that is not one network

RedCLARA is easiest to misunderstand when it is described simply as Latin America’s research network. The phrase suggests a single operator with direct control from the international backbone down to a university socket. The actual architecture is a federation. A researcher normally reaches RedCLARA through an institutional network and then through a national research and education network, or NREN. RedCLARA supplies the regional and intercontinental layer that allows those separately governed systems to function as part of a larger scientific environment.

That distinction determines both the organisation’s value and its limits. Brazil’s RNP, Chile’s REUNA, Colombia’s RENATA, Ecuador’s CEDIA, Mexico’s CUDI, Costa Rica’s RedCONARE and Uruguay’s RAU retain national governance, domestic infrastructure, institutional relationships and their own operational responsibilities. RedCLARA does not replace them. It connects them, represents their shared requirements, coordinates capacity and services, and provides a route into peer research networks such as GÉANT, Internet2 and CANARIE.

A typical end-to-end path may therefore cross a campus LAN, an institutional firewall, a national access circuit, the NREN backbone, RedCLARA, an international peer, a destination NREN and a remote research facility. A regional 100 Gbps path cannot compensate for a congested campus edge, a slow national link, misconfigured storage or restrictive policy at the destination. Conversely, a national network cannot by itself negotiate every intercontinental route, identity relationship and collaborative service required by modern science.

The useful description is that RedCLARA is a shared coordination and infrastructure layer. “Shared” refers to governance, engineering, long-term capacity rights and common services, not ownership of every fibre, router, point of presence or cable segment drawn on a regional map. This is not a semantic caveat. It is the central mechanism through which the organisation converts many national systems and commercial infrastructure contracts into a regional public-interest network.

Why Latin America needed a regional layer

The founding problem was fragmentation. By the early 2000s, several Latin American countries had national or emerging academic networks, but they did not have a durable continental organisation capable of aggregating demand, operating a regional backbone or negotiating as an equal with networks in Europe and North America. International circuits were expensive, domestic telecommunications markets varied widely, and commercial routes were designed around carrier economics rather than the requirements of distributed science.

That geography imposed practical penalties. A university might have strong domestic connectivity yet depend on expensive and indirect international transit for collaboration with another Latin American institution. Traffic between Europe and Latin America could be routed through North America because the available commercial cable and interconnection structure made that path easier to buy. The result was not only latency. It was weak bargaining power, limited route diversity and a dependence on services whose priorities were set outside the research community.

A centralised continental carrier would not have been a realistic solution. The region contained large differences in NREN maturity, public funding, regulation, fibre availability, university systems and international landing points. National organisations needed to retain their legal and operational autonomy. The viable model was therefore a federation: domestic networks would keep their national roles, while a regional association would coordinate the common layer and engage with global peers.

This design also created political and institutional value. One NREN negotiating alone might struggle to influence an international research programme or justify a specialised route. A regional organisation could aggregate scientific use cases, pool contributions, combine grants with national assets and present a coherent counterpart to the European Commission, GÉANT, development banks, carriers and cable operators. RedCLARA was created to turn that collective need into an institution and a running network.

Toledo, Valle de Bravo and the legal association

The institutional idea took shape in 2002. Latin American NREN representatives discussed the feasibility of a regional network during a CAESAR meeting in Toledo, Spain, and continued the work through meetings in Rio de Janeiro, Buenos Aires, Santiago and Mexico. Those discussions had to solve more than topology. They had to define who would participate, how a regional body would relate to national networks, how decisions would be made and how an externally funded project could become a durable organisation.

The legal chronology contains three dates that should be preserved rather than forced into one simplified founding day. RedCLARA’s history says the statutes were signed in Valle de Bravo, Mexico, on 9 June 2003. An Organization of American States civil-society record uses 10 June 2003 as the establishment date. RedCLARA also says its legal existence was recognised under Uruguayan law on 23 December 2003. The public evidence supports treating these as related but distinct events: the constituent agreement, an administrative establishment record and later legal recognition in Uruguay.

The organisation is best described as a nonprofit international civil association recognised under Uruguayan law and headquartered in Montevideo. RedCLARA sometimes uses the English phrase “international law organisation,” but the available statutes and registration material do not establish that it was created by treaty or has the legal personality of an intergovernmental organisation. It is a member association with regional objectives, not a supranational regulator and not a government agency.

Its current statutes were approved in November 2019 and published in January 2020. They establish the General Assembly as the highest governing body, a Board to direct the association, a Fiscal Committee for oversight, a Technical Commission that brings together NREN expertise and an Executive Directorate responsible for implementation. This legal structure matters because the network’s legitimacy comes from association and cooperation, not from equity ownership or territorial authority.

Seven members, eleven connected countries and a broader mission

RedCLARA’s current public member directory lists seven NRENs: RNP, RENATA, RedCONARE, REUNA, CEDIA, CUDI and RAU. Its current advanced-connectivity page, however, says the network connects eleven countries: Argentina, Brazil, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Mexico, Paraguay and Uruguay. Older material refers to thirteen countries. These figures are not interchangeable.

Legal membership determines governance rights, obligations and participation in the association. Operational connectivity can exist through a partner, transitional arrangement or project even when the relevant national network is not a current full voting member. Historical country counts may reflect earlier memberships or older programme footprints. The correct account therefore distinguishes seven current legal NREN members from eleven currently connected countries, while treating older totals as dated evidence rather than current fact.

This difference creates a strategic tension. RedCLARA seeks to serve a region larger than the association that formally governs it. A country may benefit from connectivity or a BELLA II programme while lacking an equal place in the General Assembly. Conversely, the seven members carry a disproportionate share of governance, technical work and potentially recurring financial responsibility. The gap is not proof of institutional failure, but it affects legitimacy, cost sharing and the ability to build a truly regional platform.

The current Board for the 2025–2027 term reflects the active legal membership. José Palacios of REUNA serves as President; Eduardo Grizendi of RNP as Vice President; Juan Pablo Carvallo of CEDIA as Treasurer; Carlos Gamboa of RedCONARE as Secretary; and Eduardo Grampín of RAU as a Board member. Luis Eliécer Cadenas is the Executive Director. The structure gives national networks formal control, but the public record does not provide recent Assembly minutes, voting weights, current dues or a complete account of how connected nonmembers participate in major decisions.

ALICE supplied the start-up capital for a regional network

RedCLARA’s legal formation and its first operating network were inseparable from ALICE, the Europe–Latin America Interconnection project. The European Commission and DANTE signed the project contract in June 2003. The programme had a budget of €12.5 million, including €10 million from the European Commission and €2.5 million from Latin American partners. It ran through March 2008 after extensions.

ALICE solved several start-up problems at once. It provided capital, created a common procurement process, gave the new Latin American association an immediate operational purpose and connected the region to GÉANT. Instead of waiting for a mature membership and revenue model before building infrastructure, the project allowed the network to prove its value while the institution was still consolidating.

The first network was modest by present standards, but its importance was institutional. It established regional points of presence and production interconnection among participating NRENs. It also connected Latin America to European and North American research networks, creating technical relationships that would later expand into identity, middleware, grid computing, open science and shared projects. The operating network made the association more than a forum.

ALICE also embedded a difficult question in RedCLARA’s model: what happens when a time-limited grant funds infrastructure that must operate indefinitely? A network cannot simply disappear when a project reporting period closes. Capacity contracts, equipment, technical staff and user expectations continue. RedCLARA’s later history can be read as a sequence of attempts to extend the asset horizon—from short managed circuits to IRUs, wavelengths, optical spectrum, shared services and co-investment—while preserving the regional institution between project cycles.

ALICE2 shifted coordination to Latin America

ALICE2 began in December 2008 under an €18 million European Commission contract. The decisive institutional change was that RedCLARA became the coordinator rather than merely a regional participant under a European coordinating organisation. This represented a transfer of operational and programme leadership to the Latin American association.

The project focused explicitly on sustainability. It sought lower recurring costs through dark fibre, wavelengths and indefeasible rights of use; greater geographic coverage; stronger national networks; technical training; user communities and a more stable regional organisation. Sources differ on whether the project effectively concluded in 2012 or reached administrative completion in January 2013. The safest account is that operational work wound down before final administrative closure.

ALICE2 also broadened the definition of infrastructure. Grid-computing initiatives connected regional resources with international scientific collaborations. Identity and middleware projects helped institutions authenticate users and reach shared services. Training and community work recognised that a fibre route has limited value if national networks lack staff, governance or applications capable of using it.

The project therefore moved RedCLARA from a connectivity start-up toward a regional e-infrastructure organisation. It still depended on external funding, but it gained experience in coordinating large international programmes, managing technical work across national networks and turning transport into a platform for other services. That capability later became as important as the routers and circuits themselves.

From recurring circuits to long-lived capacity rights

The economics of a regional research network depend heavily on how capacity is acquired. A short managed-circuit contract provides a defined service but leaves the operator exposed to recurring price negotiations and limited control over upgrades. A long-term fibre, wavelength or spectrum right requires more complex investment and governance, yet it can give the network a longer planning horizon and allow transmission equipment to be upgraded without repurchasing every increment of bandwidth.

RedCLARA’s historical strategy increasingly favoured IRUs, dark fibre, wavelengths, optical channels and capacity swaps where circumstances allowed. The organisation did not become a cable owner in the conventional sense, but it gained durable rights that could outlive one equipment generation. Its 2015 reporting linked this approach to lower variable bandwidth costs and greater sustainability.

Long-term rights also change risk. They reduce dependence on repeated capacity purchases but increase exposure to the underlying cable, landing station, carrier, maintenance regime and contract. Spectrum on a subsea system is valuable only while the physical cable is repaired, powered, operated and commercially supported. The research network still needs optical equipment, colocation, terrestrial extensions and skilled operations.

This is why RedCLARA’s asset base cannot be understood through a simple ownership map. Some infrastructure may be owned by an NREN, some leased from a carrier, some contributed in kind, some held through an IRU and some represented by optical rights on a shared commercial cable. The regional platform is assembled from legal and technical relationships as much as from physical assets.

Before BELLA, Europe often meant North America first

The absence of a direct high-capacity research route between Europe and Latin America shaped both performance and dependency. Even when two research institutions were located on opposite sides of the South Atlantic, traffic could follow commercial systems through North America. That route was not necessarily irrational for carriers, but it added distance, concentrated transit relationships and limited the research community’s control over capacity.

North American interconnection was and remains important. Early initiatives such as WHREN/LILA strengthened routes between Latin American networks and the United States, including connectivity through Tijuana, San Diego and Pacific Wave. Miami became a major gateway toward Internet2, CANARIE and other partners. These paths supplied essential reach and remain part of the resilience architecture.

The problem was dependence rather than the existence of North American routes. A continental network needs alternatives. Direct Europe connectivity could lower latency for appropriate source-destination pairs, provide route diversity and support large scientific datasets without forcing every flow through the same geographic corridor. It could also align the physical network more closely with the long-standing institutional partnership between RedCLARA and GÉANT.

The route question became more urgent as research workloads expanded. Earth-observation imagery, genomics, telescope data, climate modelling and high-performance-computing workflows require sustained throughput and predictable transfer behaviour. BELLA was the programme that turned the direct-route ambition into a long-term physical and contractual system.

The present topology: eight nodes, several kinds of capacity

RedCLARA’s current technical description names eight principal routed nodes: Buenos Aires, Fortaleza, Porto Alegre, São Paulo, Santiago, Manta, Panama City and Miami. The layout places several different functions inside one map. South American national networks enter a regional ring; Panama City aggregates and extends Central American reach; Miami provides a principal handoff toward Internet2, CANARIE and other North American relationships; and Fortaleza is the principal Atlantic-facing node for the direct path toward GÉANT and onward research-network partners.

The public description calls the main South American ring 100 Gbps or more. Another current service page refers to links of 100, 200 and 300 Gbps. Central American infrastructure is described as a shared 20 Gbps system. These numbers are not competing answers to one question. A backbone class, an aggregated or protected link, a regional subnetwork and an optical channel can all have different capacities at the same time.

That is why the safest summary is that RedCLARA operates a 100G-class regional network with some links described at 200 or 300 Gbps, while Central American capacity is smaller. An older 10 Gbps statement remains on RedCLARA’s website, but it is inconsistent with the current topology and appears to be documentation debt left behind after upgrades. It should not be used to downgrade the current network, just as the largest published number should not be applied to every path.

Capacity also changes meaning depending on the layer. A physical optical system may support multiple 100G wavelengths. A router interface may be protected through two links. A service may be rate-limited below the underlying port. A national NREN may receive less capacity than the regional core. The institutional access circuit may be the bottleneck even when the international route has abundant headroom. Public figures become useful only when their scope is identified.

The same caution applies to distance. EllaLink’s direct Portugal–Brazil subsea span is approximately 6,000 kilometres, while RedCLARA’s 2024 annual report describes a wider BELLA infrastructure of 35,830 kilometres. The larger figure includes terrestrial and international components beyond the single cable segment. Both can be correct; neither should be presented without explaining the boundary being measured.

A topology diagram also cannot prove physical diversity. Two logical paths may share a duct, landing station, carrier facility or power dependency. The ring shape suggests routing alternatives, but resilience depends on contracts and civil works that are not visible in a public map. RedCLARA’s strategic value lies partly in coordinating those hidden layers, yet external readers still lack a route-by-route asset and diversity inventory.

AS27750, peering and the separation of traffic classes

AS27750 is RedCLARA’s public autonomous-system identity. Through it, the organisation carries or exchanges routes for member networks, global research-and-education peers, service infrastructure and selected commercial connectivity. The routing layer supports IPv4 and IPv6, multicast, direct peering and protected regional paths. It is the mechanism by which the federation becomes visible to the wider internet routing system.

Public routing databases show relationships with networks including GÉANT, Internet2, CANARIE, ESnet, RNP, REUNA, RENATA, RedCONARE, CEDIA and TENET. PeeringDB describes a selective peering policy and identifies presences in major interconnection environments. Those records are useful operational snapshots, not a complete contract map. They can show that two systems exchange routes without disclosing settlement terms, capacity commitments, protection arrangements or which party supplies the physical circuit.

Peering can improve path length, latency and transit economics, but it creates operational work. Prefix filters must be accurate. Maximum-prefix limits need to reflect legitimate change without permitting a leak to overwhelm a session. Internet Routing Registry and Resource Public Key Infrastructure data should be maintained. Asymmetric paths complicate troubleshooting. Capacity must be balanced across maintenance and failure states. A direct BGP session is therefore a continuing engineering relationship, not simply a cable plugged into an exchange.

RedCLARA also distinguishes research-and-education traffic from commercial internet traffic. That separation is important because eligible academic destinations can be reached through specialised routes while commodity traffic may follow different peers or transit providers. It can allow NRENs to aggregate commercial demand without losing the policy and performance characteristics of the research network.

The separation is logical and operational, not absolute physical isolation. The same router, fibre, rack, cable system or carrier building may support several traffic classes. RedCLARA is interconnected with the commercial internet and offers commercial-network services; it does not exist in a sealed technical universe. Its differentiation comes from eligibility, routing policy, dedicated capacity, community trust and service design.

This distinction matters for procurement and sovereignty claims. A dedicated research path can reduce dependence on the public internet for selected workloads, but it does not remove dependence on commercial fibre owners, colocation providers, equipment vendors or cable operators. Mission alignment changes who coordinates the route and why. It does not repeal the physical supply chain.

Layer 2 circuits, operations and the limits of regional responsibility

RedCLARA’s service portfolio includes dedicated Layer 2 virtual circuits. A VLAN can make laboratories, data centres or research instruments in different countries behave as though they share one Ethernet environment. That can be useful for moving scientific data, extending a private cloud, connecting an observatory to compute resources or supporting a temporary experiment that should not rely on ordinary routed internet paths.

Layer 2 reach also extends a failure domain. Broadcast storms, spanning-tree interactions, unknown-unicast flooding, weak segmentation and MTU mismatch can travel farther than local engineers expect. VLAN identifiers and handoffs must be coordinated across RedCLARA, national networks and institutional infrastructure. A regional circuit does not transfer every design responsibility to the regional provider.

The same division of responsibility appears throughout RedCLARA’s operations. Its Network Engineering Group plans architecture, capacity and security evolution. The Network Operations Centre monitors the network and coordinates incidents. The Service Engineering Group implements and operates value-added services. National NRENs manage their own access environments, and institutions manage campuses, applications and users.

That layered model is appropriate for a federation, but it makes service measurement difficult. RedCLARA has published claims about annual traffic and monthly availability, including an undated 7.7-petabyte figure and a 100% average access-availability statement. The pages do not disclose a current methodology, measurement points, treatment of planned maintenance or whether national and institutional access is included. The claims may describe a valid historical scope, but they are not an independent current benchmark.

A professional operator evaluating RedCLARA therefore needs more than a headline availability number. It needs the service demarcation, maintenance process, escalation path, route protection, incident history, restoration target and responsibilities at every handoff. The regional organisation’s lack of a public longitudinal incident database does not prove poor reliability. It means external users cannot independently assess the pattern of fibre cuts, route leaks, packet loss, capacity events and repair times.

The practical strength of RedCLARA is coordination across boundaries that no single participant controls. The operational risk is the same thing. Every service crosses organisations with different budgets, tools and maturity. The network succeeds when those boundaries are explicit and rehearsed; it becomes fragile when a shared service is marketed as though one operator controls the complete path.

BELLA II: expansion through co-investment, not a completed map

RedCLARA and the European Commission signed BELLA II on 14 December 2022. The programme was designed for 48 months, with a €13 million European Commission contribution toward a €28 million plan. RedCLARA was expected to mobilise at least €15 million from governments, development institutions, private companies, banks and other partners. The structure makes clear that the second BELLA phase is not simply another cable purchase financed by one donor.

Its connectivity objective is to strengthen or connect Peru, Costa Rica, El Salvador, Guatemala and Honduras, with possible extension toward Mexico, the Caribbean, Paraguay, Uruguay, Bolivia and Belize. The 2025 competitive dialogue sought a system with a minimum life of 15 years, at least 20 Gbps for Central American segments and at least 100 Gbps for Peru, with room to scale. Those are procurement requirements and planning targets. They are not proof that every route had entered production by the August 2026 research cutoff.

The distinction is especially important because infrastructure programmes generate many kinds of announcement before commissioning. A strategic dialogue shows political interest. A memorandum of understanding defines cooperation. A tender or competitive dialogue explores technical and commercial options. A contract identifies obligations. Construction creates physical progress. Acceptance testing and service activation establish an operational link. Collapsing these stages would turn a future network into a present fact.

BELLA II’s procurement method reflects the diversity of the region. One route may be best served through an IRU on existing fibre. Another may require a capacity swap, co-construction, a managed service or a combination of national and international assets. RedCLARA is asking carriers, infrastructure owners, governments and financiers to help define a durable system rather than responding to one rigid circuit specification.

That flexibility can attract investment, but it also complicates accountability. Co-investment requires clarity about who funds construction, who owns the resulting asset, who maintains it, which capacity is reserved for research, what happens after the grant and how a country joins the service. A 15-year design is meaningful only when maintenance and operating finance are as durable as the initial capacity right.

BELLA II should therefore be judged by commissioned routes, enforceable capacity rights and post-project operating arrangements—not by the number of dialogues or target countries named. At the cutoff, the programme represented a credible and active expansion pathway. It did not yet justify drawing every proposed country as a completed part of RedCLARA’s production backbone.

From a transport network to a regional research platform

The first generation of RedCLARA proved that national research networks could exchange traffic through a regional backbone. The later organisation has tried to answer a more demanding question: what shared capabilities become possible once that backbone exists? Its portfolio now includes identity, file transfer, collaboration environments, training, high-performance-computing access, open-science services, cybersecurity coordination and thematic communities.

This shift is not a departure from networking. It is an attempt to increase the value of the network. A high-capacity path has limited impact if researchers cannot authenticate to a service, discover a computing facility, move a large dataset, find partners or obtain support for a cross-border workflow. The application and community layers turn raw capacity into usable research infrastructure.

BELLA II formalises that platform logic. Its activities include innovation hubs, testbeds, open calls, training, strategic dialogues and mechanisms for matching projects with funders or partners. The 2024 agrifood Ideathon, which involved 125 experts from 28 countries, showed how RedCLARA could convene a regional problem-solving process. Initial high-performance-computing testbeds created a route from training and prototyping toward shared computational resources.

A testbed is different from a production service. It normally has a selected cohort, bounded capacity, expert support and an experimental objective. It can reveal technical and organisational barriers before a region commits to permanent infrastructure. It can also create misleading expectations if a successful demonstration is presented as universal availability or commercial maturity.

RedCLARA’s opportunity is to become a service-and-testbed federation rather than a transport-only network. That would make the regional layer relevant to researchers who never think about BGP or optical channels. It would also increase operational complexity. Identity, storage, compute, notebooks, collaboration tools and sensitive data introduce security, privacy, support and lifecycle responsibilities that a backbone operator can avoid only when it remains narrowly focused on transport.

The organisation is therefore moving up the stack without owning every component. National networks, SCALAC members, universities, cloud providers and project partners continue to operate the underlying resources. RedCLARA’s role is often to integrate, select, coordinate and provide a regional access path. The platform is federated for the same reason the network is federated: the region’s capabilities are distributed, and no single institution can efficiently reproduce them all.

BIO+IA and the discipline of early-adopter evidence

The BIO+IA Early Adopters programme is one of the clearest examples of RedCLARA’s emerging platform role. Opened in 2026 under BELLA II, it selected five projects from six countries for training, expert support and access to an AI-and-bioinformatics test environment. The projects addressed breast-cancer biology, molecular screening, antimicrobial or mercury resistance, liver-disease drug repurposing and translational biomedicine.

At the July 2026 Demo Day, participating teams reported how the environment helped them search, integrate and interpret biological information. That is meaningful evidence of a facilitated research workflow. It shows that a regional programme can combine network access, specialised tools and expert guidance to help distributed teams work on data-intensive biomedical problems.

The evidence does not establish clinical effectiveness. A model that identifies a candidate molecule has not shown that a treatment works. A bioinformatics pipeline that accelerates analysis has not completed laboratory validation, regulatory review or deployment in a health system. The programme selected a limited group rather than opening an unlimited regional service. Its reported outcomes come from the participants and organisers, not an independent clinical trial.

These qualifications do not diminish the programme. They identify the correct unit of achievement. RedCLARA demonstrated that its infrastructure can support a structured regional experiment at the intersection of AI, biology and advanced networking. It reduced some barriers to tools, data and expertise for participating teams. It created relationships among researchers in several countries and produced evidence that selected workflows could be performed through a shared environment.

The next test is continuity. A demonstration becomes infrastructure when teams can return to it, new cohorts can enter, data governance is clear, compute and storage are funded, and successful workflows can move into national or institutional systems. Usage, scientific outputs, independent evaluation and post-pilot financing will reveal whether BIO+IA becomes a recurring service model or remains a valuable time-limited programme.

The larger lesson is that RedCLARA should measure innovation programmes by the distance between participation and durable capability. Ideathons count people. Testbeds count experiments. Infrastructure must also count maintained environments, repeat users, governed data, trained operators and projects that continue after the grant or event has ended.

Identity as infrastructure: FIEL, eduGAIN and eduroam

A network path is useless to a researcher who cannot prove identity to the service at the other end. RedCLARA’s work on identity federation addresses that control plane. FIEL supports national identity federations, while eduGAIN allows participating federations to exchange metadata and give users access to international services through their home institutions. eduroam provides a related roaming model for network access.

These systems do not create one regional password database. In a typical federated login, a service redirects the user toward the home institution or federation. The home institution authenticates the person and returns a signed assertion or selected attributes. The service then applies its own access policy. RedCLARA strengthens federation capability and connects national systems to the wider trust fabric; it does not normally receive every user’s password.

The architecture distributes both trust and risk. A university can keep responsibility for its users rather than handing credentials to every external service. A researcher can use one institutional identity across repositories, computing platforms and collaboration tools. Yet assurance varies among institutions. Metadata may be stale, attributes may have different meanings and a compromised identity provider or signing key can affect relying services beyond one campus.

eduGAIN solves federation discovery and metadata exchange, not every legal and policy question. Services still decide which institutions and assurance levels they accept. Institutions must protect keys, manage account lifecycle and respond to incidents. Cross-border use raises questions about consent, attribute minimisation, privacy and the legal basis for transferring identity information.

eduroam follows a comparable division of responsibility. A visiting institution supplies network access and forwards the authentication request through a national or regional RADIUS hierarchy to the user’s home institution. The home institution validates the credential. Security depends on correct device configuration, certificate validation, the home identity system and local network controls.

RedCLARA’s identity role matters because federated access is a form of infrastructure continuity. A supercomputer, repository or testbed can be reachable at the IP layer and still be unavailable to the intended user if the trust path fails. By coordinating identity alongside connectivity, RedCLARA addresses a failure mode that a transport-only network would leave outside its scope.

MiLab, eNVIO and the collaborative research layer

MiLab combines several open-source tools behind federated access. Its components include a Dataverse-based data environment, a GitLab-based development environment, a Mattermost-based communication space and notebook-style computational resources associated with JupyterHub, Python and R. Together they support a research group from early communication and code development through analysis, data curation and publication.

The design is valuable because distributed teams often assemble these functions from unrelated services with different accounts, data policies and support models. A regional workspace can reduce that fragmentation and create a common environment for projects whose members are spread across institutions and countries. RedCLARA has documented research use, including collaboration around geographically distributed data.

MiLab should not be described as a universal regional cloud. It does not automatically provide large-scale high-performance computing, unlimited storage, sensitive-data certification or permanent legal preservation. The exact capacity, service level and hosting arrangements can change. The environment is a collaboration layer whose capabilities must be matched to the project’s security, data and performance requirements.

eNVIO addresses another practical barrier: moving large files through academic infrastructure. Email and consumer file-sharing systems are poorly suited to many scientific datasets. A purpose-built transfer service can improve usability and keep research exchange inside a community-oriented service environment. Its actual value depends on file limits, retention, encryption, user support and integration with national networks—metrics that are not fully published.

RedCLARA also operates regional training platforms and a learning-management environment developed with CEDIA. These services convert specialist knowledge into courses, enrolment, assessment and certification. Training is essential because shared infrastructure does not create skilled staff automatically. At the same time, a course cannot replace funded positions, production experience or institutional responsibility.

The collaborative layer illustrates RedCLARA’s operating model in miniature. The organisation integrates tools and access, national networks bring users, institutions contribute identity and project context, and researchers supply the work. The platform succeeds when those responsibilities are clear. It fails when a regional service is expected to compensate for every local gap in staffing, data governance or application support.

Ventanilla Abierta and navigation through a fragmented ecosystem

Regional research infrastructure can be difficult to discover even when it exists. A scientist may know that high-performance computing, storage, identity or specialist networks are available somewhere in Latin America without knowing which organisation operates them, whether the project is eligible or how to begin an application. Ventanilla Abierta was created as a one-stop advisory layer for that problem.

The service does not own every resource to which it directs a user. It helps translate a research requirement into a path through the ecosystem: connectivity through a national network, computing through SCALAC or another facility, data and storage services, identity federation, an international partner resource or a project and funding route. Its value is coordination and discoverability.

That makes Ventanilla Abierta a form of navigation infrastructure. Physical capacity is often underused not because there is no demand, but because institutional interfaces are fragmented. A researcher may abandon a regional option when the process requires separate knowledge of several NRENs, programme rules and technical teams. A guided front door can reduce that transaction cost.

RedCLARA’s Funds and Partners service addresses a related problem. Research collaborations need calls for proposals, complementary institutions and financing as well as network access. Connecting a technical need with a partner or funder can turn a latent idea into a project capable of using the infrastructure. The service therefore sits between information, business development and public-interest programme coordination.

The limitation is authority. RedCLARA can advise and connect, but it cannot guarantee another organisation’s capacity, eligibility decision, budget or schedule. A referral is not a reservation. A promising funding call is not an award. An introduction is not a consortium agreement. The service must be judged by successful handoffs and sustained projects, not only enquiries received.

This layer is strategically important because it reduces the distance between the regional network and the end user. National NRENs remain the principal institutional channel, but a regional navigator can make cross-border resources legible. In a federation, discoverability is part of capacity: an unused supercomputer or identity service contributes little to regional science if potential users cannot find a trustworthy route to it.

LA Referencia, SCALAC and the shared knowledge-and-compute chain

RedCLARA’s open-science and high-performance-computing relationships show how transport becomes an ecosystem. LA Referencia aggregates metadata and research outputs from national repository systems. SCALAC links advanced-computing organisations and supports regional collaboration around high-performance resources. These are distinct alliances rather than RedCLARA subsidiaries, but the regional network and coordinating institution help them operate across borders.

The open-science chain begins at the institution or national repository. Researchers deposit publications or datasets under local policies. National systems organise metadata and access. LA Referencia creates regional discovery and interoperability, while relationships with initiatives such as OpenAIRE connect Latin American outputs to a wider international environment. RedCLARA contributes connectivity, project coordination and institutional linkage; it does not own every repository or research output.

Open science depends on more than bandwidth. Metadata quality, persistent identifiers, licences, repository governance, preservation and researcher incentives determine whether an output can actually be found and reused. A fast network can move a dataset but cannot make an undocumented file reproducible. The regional infrastructure is strongest when transport and information policy are designed together.

SCALAC addresses a comparable problem in computing. Latin America’s high-performance resources are distributed among universities, national centres and other institutions. A regional alliance can make those resources visible, coordinate expertise and enable testbeds or training for projects that would be difficult to support inside one country. CARLA-related activity and BELLA II testbeds have extended this cooperation into practical access and capacity building.

A regional HPC federation is not one supercomputer. Facilities may use different schedulers, software stacks, access policies, queues, storage systems and accounting models. Data has to be staged, users authenticated and allocations governed. Sensitive workloads may face cross-border restrictions. The network removes one bottleneck while exposing the need to coordinate the others.

The strategic benefit is optionality. A country that cannot finance every category of specialised compute may still participate in regional workflows. A facility with spare capacity can gain collaborators. A research team can prototype before seeking a larger national allocation. The risk is that project-based access remains episodic and depends on personal relationships rather than becoming a predictable service with published eligibility, capacity and cost-recovery rules.

Health, Earth observation and communities that give the network purpose

RUTE-ALC brings together university telemedicine and digital-health communities across the region. Its value lies in connecting specialists, educators and institutions that face similar challenges but operate in different national health systems. Potential uses include specialist consultation, medical education, image exchange, standards work, emergency collaboration and digital-health governance.

A research network can support these activities through reliable paths, identity and collaboration tools. It does not become a healthcare provider. Clinical privacy, medical-device regulation, professional responsibility and patient consent remain with health institutions and public authorities. A high-capacity circuit is one condition for telemedicine, not a complete clinical service.

Copernicus Academy activity in Latin America creates another sector bridge. European Earth-observation data can support agriculture, disaster response, climate adaptation, marine monitoring, water management, forestry, urban planning and energy. RedCLARA helps move data and organise regional training and innovation. The limiting factor is often analytic capability rather than access to raw imagery alone.

TICAL performs a different but complementary role. It convenes higher-education technology leaders and technical communities. The 2024 event in Rio de Janeiro drew more than 200 participants according to the annual report. Such gatherings are not infrastructure in the physical sense, but they help align procurement, security, cloud, identity and service priorities across organisations that must later interoperate.

These communities explain why a research-and-education network is not simply a discounted carrier. The organisation creates recurring relationships among people who operate similar institutions. Trust developed through events, technical commissions and working groups can accelerate incident response or project formation in ways that a bandwidth contract cannot.

Community claims still require discipline. Attendance does not prove implementation. A workshop does not create national capacity. A network connection does not demonstrate a health outcome or successful climate adaptation. RedCLARA’s contribution is strongest when it is described as enabling and coordinating the conditions under which sector institutions can act—not as owning every result produced by those communities.

EduLACSeg and the regional security trust network

Cybersecurity coordination is a natural extension of a federated network because academic institutions often face related threats across national boundaries. eduLACSeg provides a trust framework through which RedCLARA and member NRENs can share indicators, context and response experience. The value comes from seeing a campaign once and warning several countries before each institution has to discover it independently.

A July 2026 case illustrates the mechanism. Threat information originating from CERN was distributed through RedCLARA to RNP, REUNA, RENATA, RedCONARE, RAU, CUDI and CEDIA. The case is concrete evidence that the regional layer can move security intelligence among national networks. RedCLARA described eduLACSeg as the largest academic cybersecurity ecosystem in Latin America, but that superlative lacks an independently defined comparison category and should remain an organisational characterisation.

Threat sharing is not automatically safe or effective. Indicators can be wrong, stale or context dependent. Attribution may be uncertain. Personal-data and confidentiality rules differ among countries. One NREN may have a mature security team while another depends on a small group with limited tooling. A regional alert proves distribution, not uniform response.

The 2024 annual report records security training with LAC4 for 85 participants from almost 60 organisations in 18 countries, with women representing 30% of participants. It also records a Security Baseline Bootcamp conducted with RNP and GÉANT during TICAL2024. These are useful capacity signals, although training attendance should not be converted into a claim that every participating institution has implemented the same controls.

RedCLARA’s own engineering and operations groups must also protect the infrastructure they coordinate. Routing sessions, identity metadata, collaboration platforms and testbeds create different attack surfaces. Public information identifies AS27750 and operational relationships but does not provide a complete current policy for RPKI validation, ROA coverage, IRR filtering, MANRS participation or route-leak response.

The opportunity is to turn sector trust into measurable resilience: faster detection, validated warnings, coordinated exercises, shared baselines and transparent post-incident learning. The risk is that the regional network becomes a trusted channel without enough common classification, secure tooling or local response capacity. Trust can accelerate defence only when the recipients can act on what they receive.

SUBMERSE and the network as a scientific instrument

SUBMERSE extends RedCLARA’s role beyond moving scientific data. The project explored how operational telecommunications fibre can itself become a sensor. Distributed Acoustic Sensing measures changes in optical backscatter along fibre, while State of Polarisation techniques can reveal other physical disturbances. The resulting signals may help detect earthquakes, ocean waves, vessels and additional environmental phenomena.

The concept changes the value of a cable. A subsea system is normally justified by traffic capacity and resilience. Sensing adds a scientific output without requiring a separate purpose-built sensor along the entire route. The same infrastructure that carries data can generate data about the physical environment through which it passes.

RedCLARA’s relevance comes from its relationship to BELLA and the Latin American end of the direct Europe route. It can connect regional infrastructure and users with a European research consortium and help make cable access possible. It did not develop every sensing algorithm, operate every test location or own the whole EllaLink system. The project involved 24 partners, and its achievements must remain collaborative.

The principal funded phase reached a final dissemination event in February 2026. That means SUBMERSE should not be described as an indefinitely open programme without qualification. Its technical outputs and follow-on ambitions remain relevant, and the fibre-sensing model could continue through later projects, but completion of the grant phase is part of the record.

Sensing creates new operational questions. An optical experiment must not impair communications service. Data volumes can be large. Access to landing stations and fibre needs consent from infrastructure owners. Environmental signals may have security or commercial sensitivity. Scientific use also requires calibration, data stewardship and long-term analysis capacity.

For RedCLARA, the strategic significance is larger than one project. SUBMERSE demonstrates that control of long-lived network relationships can create options beyond connectivity. A regional research network may become a platform for geophysical observation, environmental science or infrastructure monitoring. That option has value only if operational safeguards, rights and funding survive after the demonstration.

An ecosystem of partners, not a collection of subsidiaries

RedCLARA’s relationship map is broad: national NRENs, the European Commission, GÉANT, Internet2, CANARIE, development banks, EllaLink, carriers, SCALAC, LA Referencia, RUTE-ALC, LNet, Copernicus Academy LAC and the SUBMERSE consortium. The breadth can create an inaccurate impression that the organisation owns or controls every initiative around it.

The evidence supports a different model. Many of these entities have separate governance and specialised mandates. LA Referencia runs an open-repository ecosystem. SCALAC coordinates advanced computing. RUTE-ALC is a health community. LNet and LACChain-related structures have their own evolving governance. EllaLink is a commercial cable system. RedCLARA contributes connectivity, coordination, project administration, convening and sometimes shared services, but partnership is not ownership.

This distributed structure is a strength. It allows expertise to remain with the institutions that possess it and reduces the need for one regional body to become a universal provider. It also complicates accountability. A user may encounter a RedCLARA-branded programme whose underlying resource is controlled by a national centre, vendor or partner. Eligibility, support and data policy may vary.

The organisation’s value therefore lies in the interfaces it creates among institutions. It can turn a collection of networks, laboratories, repositories and funding programmes into a more navigable regional system. The quality of that system depends on clear role descriptions, transparent service handoffs and honest attribution of outcomes. Overclaiming ownership would weaken, rather than strengthen, the cooperative model.

Finance is built around membership, projects and capacity rights

RedCLARA has no verified shareholders, market valuation or distributable profit. Its statutes permit member fees and other resources, and current joining material says NREN members pay annual dues for governance, service and project benefits. The current fee schedule is not public.

The organisation’s largest visible financial figures belong to programmes, not to RedCLARA’s annual accounts. ALICE had a €12.5 million project budget; ALICE2 an €18 million European Commission contract; BELLA-S1 a €12.4447 million action with €10 million in EU support; and BELLA II a €28 million plan with €13 million committed by the Commission. These numbers cannot be added and called revenue. They span different years, legal actions, partners, assets and in-kind contributions.

A large share of infrastructure value may not appear as cash passing through RedCLARA. A national network can contribute fibre, equipment, staff or colocation. A carrier may swap capacity or co-invest. A project can secure spectrum rights whose value extends for years. The economic model is therefore a combination of fees, grants, services, cooperation agreements and shared assets.

This model has delivered more than two decades of continuity, repeated EU project eligibility and strategic long-term capacity. It also exposes the organisation to grant cycles, national co-finance, currency, carrier contracts and unequal member ability to pay. No current audited organisation-level income statement, balance sheet, reserve figure or donor-concentration analysis was found. The 2024 annual report is rich in programmes but not a substitute for audited accounts. Financial transparency is therefore one of the clearest gaps in an otherwise well-documented institutional history.

The last mile is national, institutional and unequal

A continental backbone can be technically impressive while its benefits remain uneven. Latin American and Caribbean countries differ in domestic fibre, NREN funding, government support, engineering staff, cloud access, identity maturity and the quality of university last-mile connections. An institution may be nominally within a connected country yet unable to use advanced services at the required speed or assurance level.

BELLA II partly exists because the first BELLA phase did not distribute equal benefit. Peru and several Central American countries require stronger long-term routes and, in some cases, stronger national institutions. Caribbean geography adds subsea cost, multiple jurisdictions and disaster exposure. Building a regional network therefore means building or supporting NRENs, not simply landing a circuit.

This creates a difficult incentive problem. Strong members may contribute more capacity, staff and governance effort, while weaker ecosystems need more support. A purely usage-based model could exclude the countries that most need a regional platform. A purely political model could underprice the cost of maintaining high-capacity infrastructure. RedCLARA must combine solidarity with measurable obligations and avoid allowing temporary project participation to substitute indefinitely for stable membership.

The practical measure of success is not the maximum line rate on the ring. It is the number of institutions that can reliably authenticate, transfer data, reach compute, receive security support and participate in projects. That outcome depends on national policy and local investment beyond RedCLARA’s control. The regional organisation can coordinate, train and negotiate; it cannot manufacture mature domestic infrastructure by declaration.

Documentation debt obscures operational maturity

RedCLARA’s public record contains unusually rich institutional and project material, but it also contains conflicting live metrics. Seven members, eleven connected countries and older thirteen-country language appear in different contexts. Current pages describe 100, 200 and 300 Gbps links while retaining a 10 Gbps bullet. Manta appears in the current topology while older maps use Guayaquil. Forty optical channels appear in BELLA records, while a later proposal uses forty-five. Availability and traffic figures lack dates and methods.

Some differences are legitimate. A legal membership count is not a connected-country count. A 6,000-kilometre subsea span is not the same object as a 35,830-kilometre programme footprint. A grant action is not a full programme budget. Other conflicts appear to be stale content that should have been retired.

Documentation debt has operational consequences. Governments, funders and research institutions need reliable evidence to plan capacity and assess risk. An engineer can tolerate a marketing page with simplified language, but a regional infrastructure organisation should be able to publish dated topology, capacity, traffic, incident and service definitions. Without them, external users cannot distinguish growth from changed counting.

The absence of public SLA methodology, a current incident database, audited accounts, a complete asset inventory and a verified headcount does not negate the network’s existence. It limits the conclusions that can be drawn about resilience, efficiency and sustainability. RedCLARA’s next phase should treat transparency as infrastructure: a shared evidence layer that allows members and partners to make decisions on the same facts.

RedCLARA’s achievement is federation at infrastructure scale

RedCLARA’s central accomplishment is not a single cable, router or project. It is the creation of a durable regional mechanism through which autonomous national networks can pool demand, operate shared capacity, engage global peers and build services above the transport layer. ALICE created the institution and first network. ALICE2 transferred coordination and addressed sustainability. BELLA changed route geography. BELLA II is testing whether the platform can extend to countries, services and scientific workflows that remain underserved.

The organisation’s strengths are cumulative: more than twenty years of operating history, a direct Europe route, 100G-class regional capacity, established NREN governance, access to global research networks and a portfolio spanning identity, open science, HPC, health, Earth observation and cybersecurity. Its weaknesses are also cumulative: grant dependence, uneven national infrastructure, narrow legal membership, commercial supplier reliance and incomplete public operational and financial evidence.

RedCLARA is therefore neither a heroic self-contained continental carrier nor a mere project office. It is infrastructure assembled through cooperation. Its physical network depends on commercial systems, but its public-interest value comes from using those systems under a regional research mission and from creating relationships that no ordinary carrier circuit would supply.

The long-term test is whether this federation can become more durable than the projects that built it. That requires recurring member commitment, transparent operations, independent national capacity and clear decisions about which services deserve permanent support. If RedCLARA succeeds, Latin America will possess not only routes to global science but an institution capable of shaping how those routes, identities, data and computing resources work together.

The indicators that will show whether BELLA II becomes durable infrastructure

The most important near-term signal is commissioning evidence. Memoranda, dialogues and procurement requirements should be followed by named suppliers, accepted routes, live traffic and protected capacity for Peru and Central America. Capacity should be reported by service and direction, with clear distinction between initial minima, provisioned line rate, protected aggregate and usable customer bandwidth. The fifteen-year design objective matters only if operations, maintenance and upgrade rights are contractually sustainable.

Membership is the second signal. RedCLARA currently has seven publicly listed legal NREN members but an eleven-country connected footprint. Operators, funders and member networks should monitor whether target countries become durable members, partners or direct institutional participants and how each status affects voting, dues, service rights and technical obligations. A larger network without a broader recurring governance and funding base could increase operational load while leaving control concentrated.

Operational evidence is the third signal. RedCLARA should be watched for dated topology and traffic reports, an SLA methodology, incident disclosure, route-security policy, RPKI posture and measured diversity across EllaLink, Miami and alternative paths. A major cable fault or route leak would test whether the logical ring and global partnerships provide real physical and policy resilience.

Service conversion is the fourth signal. BIO+IA, HPC, MiLab, identity, eduLACSeg and open-science programmes should be assessed by repeat users, production allocations, support commitments, data-governance rules and post-grant finance rather than participation counts alone. A pilot becomes infrastructure when users can plan around it, not simply when a demo succeeds.

The final signal is finance. Publication of current audited accounts, reserves, member-fee concentration, supplier exposure and the share of unrestricted versus project-restricted funding would materially improve the organisation’s credibility. The decisive question is whether RedCLARA can maintain capacity and services between major European funding cycles without reducing weaker national networks to temporary project beneficiaries.

Evidence-supported scenarios

A successful expansion scenario would bring Peru and Central American NRENs onto long-lived, scalable routes and widen the association’s effective regional base. RedCLARA would become a more balanced continental platform, with the direct Europe path, North American gateways and Pacific-side access supporting larger scientific instruments and shared compute.

A service-federation scenario would make connectivity less visible because identity, cybersecurity, data, HPC and testbed access become the everyday products users notice. This would deepen value but also move RedCLARA into software operations, privacy and support obligations that require recurring staff and clearer service governance.

A concentrated-governance scenario would leave legal membership at seven while projects and connections expand. The organisation could remain operationally regional but institutionally dependent on a small core, creating questions about representation, cost sharing and succession.

A project-cycle scenario would see major innovations repeatedly launched through grants but struggle to survive after closeout. The backbone would continue, while higher-layer services oscillate with funding programmes. This is the most plausible downside if no durable cost model emerges.

A carrier-co-investment scenario could improve sustainability through capacity swaps, shared construction and long-term rights. It would also deepen dependence on a limited group of commercial infrastructure providers and require careful control of maintenance, exit and upgrade terms.

A network-as-instrument scenario would extend SUBMERSE methods to more regional fibre, creating environmental and geophysical data services. This remains prospective, but it could give RedCLARA a distinctive scientific role that commercial capacity alone does not provide.

Who controls the system and who carries the risk

Control in RedCLARA is distributed across several layers. Member NRENs control the association through the General Assembly and Board. The Executive Directorate and technical groups control regional implementation. National networks control domestic infrastructure and institutional access. GÉANT and other peers control their own networks. The European Commission and other funders influence programme scope through grants. Carriers, cable operators, colocation providers and equipment suppliers control critical physical dependencies.

Universities and scientific communities create demand but often lack direct governance rights at the regional level.

This structure prevents one actor from owning the whole system, yet it also creates gaps between decision power and operational consequence. A Board may approve a regional programme whose last-mile costs fall on an NREN. A funder may define a four-year objective while members inherit decades of maintenance. A carrier may own the only practical route in a country. A user may depend on a service without knowing which organisation can repair it. RedCLARA’s leadership should treat these interfaces as a primary control surface.

The incentive design should reward durable contributions rather than only project participation. Membership dues, fibre contributions, staff secondments, service operation, data stewardship and security response are different forms of value and should be made visible. Countries receiving expansion support need a credible path toward national ownership and recurring obligations. Stronger members need assurance that solidarity does not become an unlimited transfer of cost without governance accountability.

Supplier contracts should preserve route diversity, upgrade rights, auditability, exit options and continuity if a commercial counterparty changes ownership or strategy. Long-term spectrum and fibre rights reduce recurring price risk but can become stranded if landing facilities, optical equipment or maintenance relationships fail. The most consequential decisions are therefore not only how much capacity to buy, but which dependencies become difficult to reverse.

Decisions that cannot be deferred to the next grant

First, RedCLARA must decide whether it will remain primarily a connectivity federation or become a permanent operator of shared digital platforms. MiLab, identity, cybersecurity and AI testbeds create value, but each adds data, software, staffing and liability obligations. Services selected for permanence need transparent ownership, lifecycle funding, incident response and retirement rules.

Second, the association should resolve the mismatch between legal membership and operational reach. Connected countries need a clear status, voice and pathway to membership or another accountable arrangement. Leaving the gap informal may be convenient during expansion, but it creates long-term ambiguity over who governs, pays and can exit.

Third, financial transparency should become a governance instrument. Audited accounts, reserves, dues, restricted project funds and supplier concentration are not administrative details; they determine whether the network can survive a funding interruption or cable emergency. Members cannot make responsible long-term commitments without this evidence.

Fourth, RedCLARA should publish a continuity architecture. That includes route diversity, incident escalation, registry and configuration backups, identity and service recovery, cable-fault alternatives and succession for project-funded platforms. The regional network must be able to preserve essential service even when one grant, supplier, route or national institution fails.

The irreversible risk is not simply that a project misses a deadline. It is that the federation accumulates critical services and long-lived dependencies faster than it builds recurring authority, finance and operational evidence to sustain them. The irreversible opportunity is the opposite: use BELLA’s route rights, NREN governance and global partnerships to create a regional infrastructure commons that no single country could build alone. Leadership should preserve the federation’s distributed character while making responsibility explicit enough that shared ownership never becomes shared ambiguity.

The long-term significance of a shared regional backbone

RedCLARA demonstrates a form of infrastructure sovereignty based on cooperation rather than isolation. Latin American science gains greater agency not because the organisation owns every fibre, but because national networks jointly control a regional layer, hold strategic long-term capacity and participate in global research partnerships on more equal terms. Dependence on commercial infrastructure remains, yet the mission, routing relationships and upgrade decisions are no longer left entirely to commodity internet markets.

That model is valuable beyond the research sector. It shows how smaller or unevenly resourced institutions can aggregate demand without surrendering national autonomy, and how grants can be converted into rights and organisations that outlive a project. It also shows the limits of federation: coordination cannot compensate indefinitely for weak domestic networks, opaque finances or unclear accountability.

The decisive strategic choice is whether RedCLARA treats the next phase as expansion of a map or maturation of an institution. More countries and faster links will matter, but the durable asset is the system of governance, operations, identity, data, security and scientific cooperation around them. If that system becomes transparent, financially resilient and genuinely regional, RedCLARA will remain one of Latin America’s most consequential pieces of shared digital infrastructure. If it does not, the network may continue to carry traffic while its higher ambitions remain tied to the next externally funded programme.