top of page

We make the future of AI and datacenters fast, green, cost- & energy efficient

A new category of super servers for datacenters

Datacenters are at the core of every technology company. That is why the market has seen tremendous growth over the last years and is estimated to grow from $130bn today to   

in annual revenue in 2027. 

With the rise of Artificial Intelligence this market is experiencing another major boost as AI workloads can be 10-100x bigger than traditional workloads.

 

ExpectedIT creates a new category of super servers, significantly  speeding up data movement and raising efficiency so that critical growth, cost and sustainability hurdles of the industry can be overcome.  

$190bn

up-to 10x

Faster

Our technology enables up-to 10x faster AI, cloud and traditional workload jobs, by removing data-movement bottlenecks across a job's hardware devices.

often 2x more

Cost efficient

Our technology enables common job types to save 50% of hardware costs. Our technology removes data-movement bottlenecks, so that a job completes in less time and thus uses its hardware for less time.

often 2x more

Sustainable

Our technology enables common job types to save 50% on United Nations' sustainability categories such as CO2, Energy, Water and Waste. Our technology removes data-movement bottlenecks, so that a job completes in less time and thus uses less of Planet Earth's resources.

Highly

Compatible

Our technology is compatible with the vast majority of all datacenter hardware and software and allows seamless integration into existing datacenters. It is for the regular refresh of datacenter servers and leverages as-is datacenter devices.

AI's and datacenters'
cost challenge

AI services and solutions often face cost, capability and performance challenges from their datacenter hardware infrastructure. These costs, capabilities and performances are improving with datacenter's underlying chip-based devices: CPUs, GPUs, memories, accelerators, ...


However, over the past decade, on-device performance has improved 100x faster than off-device data movement. The worsening bottleneck increasingly causes chips waiting for data, which increases the time and cost of a job. This bottleneck leads to massive costs and inefficiencies for AI and more workloads and for datacenter providers.


ExpectedIT's new solution to this bottleneck enables the best use of each device's precious off-device data movement. By removing the bottleneck, ExpectedIT's new solution allows a job to complete in less time and thus be faster and use less hardware, for cost and performance improvements up to 10x.

"AI may soon require more electricity than all electric vehicles combined" WSJ headline, 15.12.2023

Global datacenters contribute 〜3% of the global man-made CO2, more than the aviation or shipping industry, says theshiftproject.org .

Datacenters with evaporative cooling in water-stressed areas take water from drought- impacted communities.

Globally more than 50m tons of e-waste were generated in 2019. 100m forecast by 2050 especially threatens citizens and nature when disposed in third-world countries.

The sustainability challenge of AI and datacenters

Our technology often allows

-50%

Cost

-50%

CO2 & Energy

10x

faster jobs

-50%

Water

-50%

Waste

Founders

CEO

Burkhard

Steinmacher-Burow

  • 21 years IBM Senior Staff

  • Blue Gene HW architect

  • 100 patents on new HW&SW

Chief Engineer

Árni

Ingimundarson

  • 6 years in the IBM POWER CPU design team

  • 11 years of leading the development of customers’ first-of-a-kind chips at Texas Instruments

CCO

Klaus

Entenmann

  • 30+ years in Financial Services

  • 10 years CEO Daimler Financial Services AG/ Daimler Mobility AG

  • Supervisory Board member Standard Chartered Bank AG

Chief Investor Relations

Alex

Schwarz

  • 6 years CEO of NIO Europe and Airbnb DACH

  • Former Management team of Paypal

  • Serial Founder & Advisor

Mission

Scalable AI and Sustainable Datacenters


At ExpectedIT, we believe that technology should help solve some of the world's biggest problems to help create a better tomorrow. Our technology will have a massive impact on the scalable and sustainable use of data in the AI era.

Contact us today if you want to 

  • join our team creating our new piece of datacenter architecture.
    We are hiring chip designers and hardware developers who want to dent the AI and datacenter universe.

  • become an early partner or customer of our technology, for a competitive advantage over your competitors.

bottom of page