Questions, answered.

It takes 12 weeks to build and deploy a pod. For large deployments (100MW+), a gradual deployment schedule may be implemented.

There are a few reasons. One is heat transfer: there is a much smaller surface area requirement for heat transfer within fluid as compared to air. Modular design is better in and of itself, but combined with immersion cooling, it is extremely efficient.

Modular also means scalable, rapid, agile, and less expensive than the traditional data centers we are building today.

We can build anywhere in the world. We can deploy anywhere there is water.

We build from sustainable materials that can be recycled and that are proven to be suitable for the subsea environment – both in terms of integrity and marine life. We use less carbon intensive steel, utilize environmentally friendly fluids and recycled materials where possible.

We place servers and sensors in each pod that enable computational workloads and the ability to monitor the environment.

Our bill of materials count is substantially lower than that of a traditional data center. The complexities of building and extending cables on land do not exist in the subsea environment, further reducing cost.

There are times where speed and agility are the only requirements, and other times where the temporary, transitional ability of a pod is what makes it attractive.

Shipping, reassembling and testing military or survey data centers can result in down times of weeks. Subsea Cloud's solution is fully self-contained and can be set up within hours of arrival.

A pod can be deployed and up and running on the seafloor anywhere between 2h – 6h upon arriving at its location.

In some deployments we do – typically when deploying in a port or very shallow water. The bulk of our deployments are accomplished with remotely operated vehicles (ROVs).

We can install in any body of water, but we tend to install in waters where there is existing power and telecommunications infrastructure.

It depends on the servers housed within. However, each pod can accommodate between 0.5MW and 1.8MW.

There are deep-sea cables – transoceanic fiber optic cables that are responsible for connecting much of the world's communication via the internet, carrying as much as 99 percent of total global internet and signal traffic.

There are unmanned underwater vehicles (UUVs) and remote operated vehicles (ROVs).

Petroleum companies place infrastructure such as pipelines, wellheads, power units, drilling and production equipment, refining and chemical equipment, and more on the seafloor.

You can schedule periodic maintenance, including server replacement, for any interval you deem necessary. It takes between 4–16 hours to get to site, pull up the required pod(s) and replace the servers in any case.

It costs significantly less. Through our strong partnerships we have made maintaining and operating our data centers far more economical.

Our approach improves the performance and reliability of servers because our fluid is less corrosive than oxygen. This results in a reduction in failures.

Subsea data centers typically do require less maintenance. These reliability advances include a reduction in corrosion and electrochemical migration, lessening of environmental contamination like dust, debris and particulates, and reduced thermal shock. Furthermore, due to the increased heat capacity of fluids, there is a significant increase in compute performance. There is also the added benefit of having no humans interact with the servers – either performing a task erroneously or bumping into cabling, etc.

Server maintenance is often done remotely. If physical maintenance is needed, we provide the client/cloud provider a window.

The lifespan of the actual data center pods is around 25 years. However, in line with a typical server lifespan, each pod would likely live on the seafloor for 3–5 years then be retrieved. Upon retrieval, new servers would be placed within and it would be redeployed.

Systems often can be managed remotely. Contingencies are built in for different clients and different use cases.

This is entirely possible. We just need to know.

We use free, passive cooling which means there is no need for harmful greenhouse-gas refrigerants, intrusive cooling towers or computer room air handlers. This means we produce no wastewater and therefore have no impact on water quality, fish or wildlife. Our thermal footprint is low due to the heat carrying capacity of liquids. We are always open to working with environmental groups to both garner their confidence and prove our benefits to the environment and planet.

We are. We completely eliminate electrically driven cooling which is a huge win for the environment. Each pod we put into the water can stop as much as 750 tons of carbon from being deployed, per pod, per year.

We use free, passive cooling which means there is no need for harmful greenhouse-gas refrigerants, intrusive cooling towers or computer room air handlers.

Data centers use more than Britain's total electricity consumption. At 3% of the global electricity supply, data centers have the same carbon footprint as the aviation industry. Unlike the aviation industry, the trajectory of data center deployment and consumption is upwards.

To limit global warming to 2015 Paris Accord levels, the world's net emissions of greenhouse gases need to drop to zero by 2050. Instead of waiting decades to steer the data center industry in the right direction, we should do it now.

Climate change doesn't affect everyone to the same degree. Within America and around the world, the poorest people are often most drastically impacted — money pays for air conditioning, CO2 offsets, safer places to live, and food as prices soar, while impoverished families are often the least to blame yet bear the worst of the impact.

We ultimately want to do our part to help repair the climate issue. We reduce the typical CO2 emitted by a traditional data center by 40%. Each pod we deploy saves between 350–750t of carbon from being deployed each year. We use no water and we have an extremely low thermal footprint. We are working with environmental agencies and activists to garner their confidence and support.

We do not use water. Our design eliminates the need for water. This is especially important given the increasing scarcity of drinking water worldwide caused by agriculture, urbanization, and climate change. Traditional data centers use millions of gallons of water per day, 57% of which is estimated to be potable. Our subsea fluid-filled pods eliminate this water use.

The units can be used for all the same things traditional data centers accommodate.

All units are AI/ML-ready. They have the highest rack density, highest performance and lowest cost, PUE and WUE on the market today. We have PPAs with many EU power providers.

Our subsea units are suited to edge deployments, but they can also be used for CDNs, PoPs, and large-scale deployments.

They can be deployed at scale. It's much better for the environment, land use and costs to deploy at scale in the marine environment.

The minimum duration is 30 days. There is no definitive upper limit to the testing period for participants.

Yes, participants will be billed for some of the key components typically involved in data center use, such as space and power. They will not be charged for cooling costs, security, or backup and recovery.

Yes, this is possible with communication.

There are multiple options in terms of the appropriate logistics. We will also accommodate the provider as requested.

The project will be powered mainly via wind power and, when required, from another critical path from other renewable-centric providers.