54 Tbps or 30? Bangladesh’s Bandwidth Forecast Sparks Debate

Business groups and the government are now effectively at odds over Bangladesh’s future international bandwidth requirements. Private submarine cable operators are forecasting a need for 54 Tbps of international bandwidth by 2030 and are calling for new cable connections. In contrast, state-owned infrastructure data and alternative mathematical analyses suggest that regardless of how rapidly total data traffic grows, a large portion of it is being carried within the country through local caching, content delivery networks (CDNs), and peering. As a result, treating national data traffic and actual international bandwidth demand as the same thing could lead to an inflated assessment of the potential crisis.

The key issue in the debate, therefore, is not simply how many Tbps of data are being consumed. It is also essential to determine how much of that data crosses international borders, how much is delivered locally, and how much of the international traffic is actually carried through submarine cables.

Where Is the 13.5 Tbps Currently in Use Coming From?

According to a recent industry presentation, Bangladesh’s international bandwidth usage reached approximately 13.5 terabits per second (Tbps) in 2026. It was just 50 gigabits per second (Gbps) in 2013, 0.76 Tbps in 2018, 1.78 Tbps in 2020, 4.2 Tbps in 2022 and 6.86 Tbps in 2024. In other words, international bandwidth usage has increased by approximately 260 times over 13 years.

The country’s two main government-owned international connections are the SEA-ME-WE-4 and SEA-ME-WE-5 submarine cables. According to relevant data, the combined capacity of the two submarine cables is approximately 7.1 to 7.2 Tbps. According to BSCCL’s own information, there is scope to increase this capacity further through technological upgrades.

However, this is where an important discrepancy in the calculations emerges.

According to industry analysts, less than 30 percent of Bangladesh’s total data traffic currently crosses international borders. Because of local caching, CDNs and peering, this proportion is gradually declining. Of this international traffic, approximately 4 Tbps is currently being carried through BSCCL’s submarine cables, according to relevant engineering analyses. A large portion of the remainder comes from India through ITC networks, a significant share of which is actually traffic delivered to local users through content providers and CDN-based networks.

In other words, national data traffic, international bandwidth usage and the actual international traffic carried through submarine cables are three different things.

The 54 Tbps Forecast

According to forecasts by private-sector operators, Bangladesh’s bandwidth demand could reach 19.1 Tbps in 2027, 27 Tbps in 2028, 54 Tbps in 2030 and 305 Tbps in 2035. A similar forecast was also presented at a recent industry workshop.

Mohammad Aminul Hakim, chief executive officer of Metacore Subcom Limited, said that Bangladesh’s digital sovereignty would be at risk and dependence on ITCs would increase unless private submarine cables are added.

Bangladesh’s Digital Sovereignty at Risk Without Private Cables      According to him, the country’s internet demand could increase several-fold by 2030. Therefore, if private-sector submarine cables or new international connections are not added, dependence on India and the risk of long-term bandwidth shortages could threaten digital sovereignty and the quality of internet services. Infrastructure diversification is essential to address this challenge. In such a situation, he said, the introduction of private cables could reduce internet prices by up to 50 percent, improve service quality and lower data latency. He also believes that adding new submarine cables could further accelerate internet usage growth.

Is a Large Share of ‘International’ Bandwidth Actually CDN Traffic?

This is the question being raised by Engineer Md. Ariful Haque, deputy general manager (Planning and Development) at Bangladesh Submarine Cable Company PLC (BSCCL).

According to him, discussions about Bangladesh’s digital future should not focus solely on how many Tbps of international bandwidth will be required. The actual sources and uses of bandwidth need to be examined separately.

This argument also has an important point of convergence with international research. TeleGeography says the role of content providers, cloud networks and other private networks is rapidly increasing in the growing use of international bandwidth. At the same time, local caching and content generation can somewhat reduce the growth of international traffic.

Where Does the Government See the Crisis?

Asked where the obstacle lies in approving licenses for private-sector operators to maintain a balance between projected bandwidth demand and supply, Posts and Telecommunications Minister Fakir Mahbub Anam said that although the companies had received individual approvals, they had not obtained prior government approval to form a consortium.

BSCCL’s own information also identifies the SMW-6 project as part of its plan to meet future demand and notes that the lifespan of the first submarine cable will end in 2030. The company has also said that work has begun to connect the country to a fourth submarine cable within the same period.

When these statistical issues were raised, the minister, at a press conference, emphasized responsible data management by bringing government databases into a ‘connected ecosystem’.

Target: 50 Tbps Capacity by 2030

Outlining the government’s future plans, Prime Minister’s Adviser for Posts, Telecommunications and Information Technology Rehan Asif Asad said Bangladesh’s data traffic is doubling every two and a half years.

After Cox’s Bazar and Kuakata, Mongla Port in Khulna Is the Next Target He said that following this trend, traffic could reach 20 Tbps by the end of 2027 and 50 Tbps by 2030. Explaining that the government is working toward this target, he noted that Cox’s Bazar and Kuakata are currently being used for submarine cable connections, while a forward-looking plan is also being considered to establish a submarine cable connection through Mongla Port in Khulna in the future. Another important figure also emerged from his remarks. He said Bangladesh’s peak internet traffic is currently approximately 12.1 Tbps and could rise to around 30 Tbps by 2030. At the same time, the government is moving toward building at least 50 Tbps of capacity. To ensure this capacity meets demand, he said at least three new cables may be required through government and private-sector initiatives within the next three to five years.

‘There Is No Alternative to Multiple Submarine Cables’

Echoing this need, ISPAB Senior Vice President Saiful Islam Siddiq said there is no alternative to multiple submarine cables.

In other words, although there is debate over the actual demand for international bandwidth, the industry’s concerns about network redundancy and resilience are separate issues and cannot simply be dismissed on the basis of demand forecasts.

TeleGeography’s 24 Percent Calculation

According to a 2026 study by global research firm TeleGeography, global used international bandwidth could grow at an annual rate of approximately 24 percent between 2025 and 2035. Content providers, cloud networks and AI-related infrastructure are among the main drivers of this growth. TeleGeography reported that networks operated by content providers accounted for approximately 75 percent of globally used international bandwidth in 2025.

In Bangladesh’s case, if the same 24 percent annual growth rate is mathematically applied for four years to the 13.5 Tbps recorded in 2026, the figure would reach approximately 32 Tbps by 2030. This is significantly lower than the private-sector forecast of 54 Tbps.

However, an important caveat is necessary. TeleGeography’s 24 percent figure represents the CAGR of global used international bandwidth, not Bangladesh’s own forecast. Applying this rate directly to Bangladesh is merely a mathematical scenario, not an independent forecast. The research firm itself notes that caching and content localization can reduce the actual requirement for international traffic.

Questions Remain Over the 85–93 Tbps Calculation

Telecommunications policy and engineering analyst Md. Munir Hasan has conducted a detailed analysis of these conflicting forecasts regarding the country’s future bandwidth demand.

Using a bottom-up mathematical model, he demonstrated that there is a theoretical gap between international traffic and total data usage.

According to his calculations, based on 85 million active subscribers in the country and data consumption per device, overall national traffic could rise to between 85 and 93 Tbps by 2030. However, approximately 70 percent of that data could be exchanged within the country through local content delivery networks, local caching and peering servers.

Data from the Internet Service Providers Association of Bangladesh (ISPAB) has also shown instances in which more than 80 percent of the country’s data traffic was handled through local cache servers. In a network assessment in 2024, the ISP sector also stated that more than 80 percent of traffic was being served through local caching.

Munir Hasan’s analysis suggests that actual international traffic crossing Bangladesh’s borders could reach a maximum of 26–28 Tbps in 2030. Excluding ITC traffic, the actual load on submarine cables would therefore stand at around 19–21 Tbps.

Even under an extremely adverse scenario—with annual growth of 40 percent and 60 percent local caching—his model shows that demand would not exceed 30 Tbps. This would remain within the approximately 41 Tbps of state-owned capacity expected to become available after the launch of SEA-ME-WE-6.

So, Where Is the Real Crisis?

This is where the fundamental difference between the calculations of businesses and the government becomes apparent.

Private-sector forecasts suggest that international bandwidth usage is rapidly heading toward 54 Tbps. According to one government estimate, peak traffic in 2030 could be around 30 Tbps, while another mathematical model suggests that international traffic crossing the border could remain between 26 and 28 Tbps after local caching is excluded.

Meanwhile, if TeleGeography’s global annual growth rate of 24 percent is applied to the 13.5 Tbps baseline, the resulting mathematical demand for 2030 would be approximately 32 Tbps.

Therefore, the gap between 54 Tbps and 30–32 Tbps is not merely a difference of opinion. It largely depends on which data is being defined as ‘international bandwidth demand’.

According to analysts, forecasting a crisis without clearly distinguishing between total domestic data, peak-hour traffic, international transit and actual submarine bandwidth imports could be misleading.

However, there is another side to the issue. Even if international bandwidth demand is lower than estimated, excessive dependence on one or two routes, submarine cable outages, maintenance, geopolitical risks, latency and insufficient route backup can pose major risks to the network. TeleGeography also views new international infrastructure, content localization and regional connectivity as important components of the network architecture of the future.

Therefore, for Bangladesh, the question is not simply, “How many Tbps will be needed by 2030?” Rather, the questions should be: How much will be actual international traffic? How much can be cached locally? How much will come through ITCs? How much will need to be carried by submarine cables? And how much of that capacity should be kept in reserve for redundancy?

Unless these four calculations are reconciled, a forecast of 54 Tbps could be exaggerated, just as concluding prematurely that 41 Tbps of capacity will be sufficient could also be risky. Experts believe that alongside expanding Bangladesh’s digital highway, ensuring alternative routes, low latency and adequate redundancy could be the long-term solution.

//DBTech/IH/NAP/MIST/OR//