SK Telecom Builds AI Infrastructure Momentum with GPUaaS and the Haein Cluster

The growing demand for artificial intelligence computing is reshaping the role of telecommunications operators. As AI models become larger and more computationally intensive, the need for high performance infrastructure has moved into sharp focus. In response, SK Telecom is positioning itself not only as a connectivity provider but also as a key supplier of AI infrastructure through its GPU-as-a-Service offering.

At the centre of this strategy is the Haein GPU cluster, one of the largest AI computing platforms in South Korea. Built around more than 1,000 GPUs from NVIDIA based on the Blackwell architecture, the platform provides the computing power required for large scale AI training and inference workloads. The cluster represents a significant step forward from the earlier infrastructure based on NVIDIA H100 GPUs and forms part of SK Telecom’s wider sovereign AI infrastructure initiative.

The Haein cluster is hosted within the company’s Gasan AI Data Center in Seoul and is designed to deliver high performance computing capacity at national scale. The system supports intensive AI workloads including the training of large language models while also providing the flexibility required for enterprises and research organisations developing their own AI applications. The platform’s architecture allows large GPU resources to be combined into a single cluster while still being dynamically allocated to different users depending on demand.

A key component enabling this flexibility is SK Telecom’s proprietary virtualisation platform known as Petasus AI Cloud. The software layer allows the large GPU cluster to be partitioned and reconfigured dynamically, enabling customers to access the exact amount of computing power they require. This capability is essential for GPU-as-a-Service platforms where workloads can vary significantly, from small development environments to large scale model training that requires hundreds of GPUs operating simultaneously.

Alongside this, the company provides operational management through its AI Cloud Manager platform. This AIOps based environment supports the full lifecycle of AI services including development, training, deployment and operational monitoring. By combining infrastructure with operational tooling, SK Telecom aims to provide a more integrated AI computing platform rather than simply raw GPU capacity.

The Haein cluster also plays an important role in South Korea’s national AI strategy. The platform has been selected to support a programme led by the Ministry of Science and ICT that focuses on strengthening the country’s AI computing infrastructure and enabling the development of competitive national AI foundation models. Through this initiative, the cluster will contribute computing resources to projects developing sovereign AI capabilities tailored to the Korean language and domestic industries.

The name of the cluster itself reflects this national perspective. Haein takes inspiration from Haeinsa Temple, which houses the historic Tripitaka Koreana, a vast collection of Buddhist scriptures recognised as a UNESCO World Heritage archive. The naming reflects the ambition to create a modern repository of digital intelligence, supporting the development of AI knowledge and capabilities within the country.

Delivering infrastructure at this scale requires a broad ecosystem of partners. SK Telecom has worked with companies including Supermicro and Penguin Solutions to design and deploy the server infrastructure and integrated AI data centre solutions required for the cluster. These collaborations enable the rapid deployment of high density GPU servers and the supporting cooling, power and networking systems necessary to run large scale AI workloads.

The industry has already taken notice of the platform. The Haein GPU cluster was recognised at the MWC Barcelona 2026, where SK Telecom received the Best Cloud Solution award at the GSMA Global Mobile Awards. The recognition reflects the company’s continued progress in cloud and AI infrastructure development and marks the third consecutive year that its cloud related technologies have been acknowledged in this category.

For telecoms infrastructure professionals, SK Telecom’s GPU-as-a-Service strategy illustrates how operators are expanding beyond traditional connectivity services. By building large scale AI computing platforms inside their data centre footprint, operators can leverage existing strengths in infrastructure, power management and network integration to participate in the rapidly growing AI economy.

As AI adoption accelerates across industries, the demand for scalable computing infrastructure will continue to grow. With platforms such as the Haein cluster and its GPUaaS offering, SK Telecom is positioning its network and data centre assets as part of the core infrastructure supporting the next generation of AI innovation.

Related Posts: 

• Telecoms Infrastructure Blog: SK Telecom's Vision for Future Telco Infrastructure in the AI Era
• Operator Watch Blog: SK Telecom on 'Applying AI to Transform Customer Experience'
• Free 6G Training - SK Telecom 6G White Paper: View on Future AI Telco Infrastructure
• Free 6G Training: SK Telecom Releases 6G White Paper Looking at Lessons Learned in 5G and Way Forward with 6G
• Free 6G Training: SK Telecom Claims ‘Bare Metal-Based Cloud-Native 5G Core’ as Foundation for 6G Evolution
• Telecoms Infrastructure Blog: SK Telecom’s 5G MEC Status and Plan 

Data Centers At Meta: Heterogeneous Integration Driven By AI/ML And Network Applications

Last year, tech giants including Intel, Meta, Arm, Google Cloud, AMD, Qualcomm, TSMC and ASE formed the Chiplet consortium. A news article in Fierce Electronics said:

Several giant tech companies have joined hands to promote an open standard for chip components called chiplets and how they are crammed together in system-on-chip (SoC) designs deemed critical to a variety of future handheld and high-performance computers that power AI applications and much more.

The open standard, called Universal Chiplet Interconnect Express (UCIe), has been developed by Intel and clearly benefits Intel’s integrated device manufacturer (IDM) strategy as it builds new chip fabs in Arizona and Ohio and elsewhere outside the U.S.  Intel has been a prominent voice in the push to expand chip manufacturing outside of Taiwan and the rest of Asia where it is heavily focused today.

Intel has donated its UCIe standard to founding members in a new consortium that includes Intel along with Advanced Semiconductor Engineering, Taiwan Semiconductor Manufacturing Co., AMD, Arm, Google Cloud, Meta, Microsoft, Samsung and Qualcomm. The founders have already ratified UCIe 1.0 which covers the die-to-die physical layer, die-to-die protocols and software stacks which leverage the existing PCI Express (PCIe) and Compute Express Link (CXL) industry standards.

Ravi Agarwal, a technical sourcing manager at the Facebook/Meta Infrastructure group is responsible for driving advanced packaging architectures and foundry for both networking and AI/ML compute applications to meet Facebook’s future workloads. He is driving Chiplet Business Workstream in Open Domain-Specific Architecture (ODSA) Sub-Project within the Open Compute Project (OCP), working with ecosystem partners to enable a Chiplet marketplace. 

In a talk delivered for the IEEE Electronics Packaging Society (EPS) SFBA, he focused on heterogeneous integration for Artificial Intelligence, Machine Learning and network applications at Meta Infrastructure, and discussed implications for packaging and system-level considerations. In the talk he also shared some of the advanced packaging (chiplet) initiatives in which Meta is participating to develop an open ecosystem.

The talk is embedded below:

While the slides of this talk is not available, you can see slides of another talk he delivered here.

Related Posts: 

• Telecoms Infrastructure Blog: The Role of Border Gateway Protocol (BGP) in Making Sure Internet Works
• Telecoms Infrastructure Blog: Facebook's Fiber Deploying Robot
• Telecoms Infrastructure Blog: 'Story of the Edge' by Akamai 

'Story of the Edge' by Akamai

People share some useful stuff on social networks that I bookmark to visit and write later. This was shared by John Francis Nolan nine months back or so. I have just managed to watch it and definitely worth a share here.

Prof. Ramesh K. Sitaraman is currently a Distinguished University Professor and the Associate Dean for Educational Programs and Teaching in the College of Information and Computer Sciences at the University of Massachusetts at Amherst.  He is best known for pioneering content delivery networks (CDNs) and edge computing services that currently deliver much of the world’s web, videos, edge applications, and online services. As a principal architect, he helped create the Akamai network, the world’s first major content delivery network (CDN) and edge computing service. He retains a part-time role as Akamai’s Chief Consulting Scientist. 

Late last year he delivered a talk 'Living on the Edge for a Quarter Century: An Akamai Retrospective' on The Networking Channel. The following is the abstract of the talk:

As Akamai and the creation of the “edge” turn 25, we look back at the key role that the edge has played in the evolution of internet services. The story of the edge starts in the late 1990s when servers were deployed worldwide to provide content delivery services for web pages and videos. The quest to move dynamic content and application logic closer to users created the first edge computing services a few years later. The growth of the edge that now spans thousands of locations has dramatically increased the scope and importance of services that critically rely on it. This has created new challenges in operating the edge at scale and securing it from sophisticated attacks. Further, as the carbon footprint of the edge grows rapidly, reimagining a sustainable “zero-carbon” edge that is powered by renewable energy poses a key direction for future research.

His talk is embedded below and the PDF copy of his presentation is available here.

You can view the author's publication page here.

Related Posts: 

• The 3G4G Blog - Webinar: Where Edge Meets Cloud by Dean Bubley
• The 3G4G Blog: Transitioning from Cloud-native to Edge-Native Infrastructure
• Telecoms Infrastructure Blog: Operator Cloud Infrastructure and Innovation Strategy
• Telecoms Infrastructure Blog: ZTE explains 5G CDN in a new White Paper
• Telecoms Infrastructure Blog: AWS Edge to Power Private Networks and Industry 4.0
• Telecoms Infrastructure Blog: The Provider (Telecom, Cloud) and User Open Edge: Innovation with No Vendor Lock-in
• Telecoms Infrastructure Blog: Building the telco edge with or without hyperscalers! 

AWS for Public and Private 5G Networks

I wrote about AWS Edge being used to power Private Networks and Industry 4.0 back in March. Since then we had this big announcement from DISH and AWS about formation of 'Strategic Collaboration to Reinvent 5G Connectivity and Innovation'.

It talks about how the new US operator, DISH, will leverage AWS infrastructure and services to build a cloud-based, 5G Open Radio Access Network (O-RAN) that delivers consistent, cost-effective performance from core to the edge. 

Netmanias has done an awesome job of explaining how AWS will be used in the Dish network and compares it with the Rakuten Virtualised network. Reproducing the original from them below.

In addition, they have also done a fantastic job of explaining how different operators are planning to use AWS in their Networks. 

You can read more details for each of the operators below:

• AWS and Verizon Expand 5G Collaboration with Private MEC Solution
• AWS and Vodafone Business Bring Edge Computing Closer to Organizations in Europe
• Announcing the first AWS Wavelength Zone in South Korea on SK Telecom (SKT)’s 5G network
• KDDI To Launch "AWS Wavelength" On December 16, Offering Ultra Low Latency on the 5G Network Edge
• Singtel and Optus expand 5G ecosystems with AWS for 5G edge computing
• Telefónica Germany / O2 builds new 5G core network in the cloud
• DISH and AWS Form Strategic Collaboration to Reinvent 5G Connectivity and Innovation
• Bell Canada teams up with AWS for edge computing

Strategies and status of AWS entering Mobile Operators' 5G network: MEC on AWS → 5G Core on AWS → 5G RAN on AWS - https://t.co/imm61Q9TdE#Free5Gtraining #3G4G5G #Netmanias #5G #5GNetwork #Private5G #PrivateNetworks #AWS #MEC #Outpost #Wavelength #5GRAN #vRAN #5GCore #5GC #5GS pic.twitter.com/OpevhJABN5

— 5G Training (@5Gtraining) June 7, 2021

Let us know what you think about the operator strategy of moving to AWS for something or other.

Related Posts: 

• Telecoms Infrastructure Blog: AWS Edge to Power Private Networks and Industry 4.0
•  Operator Watch Blog: Rakuten details its Open RAN and Innovation Journey
• Telecoms Infrastructure Blog: Rakuten details how Cloud will make Telecommunications Industry Irrelevant
• The 3G4G Blog: 5G Private and Non-Public Network (NPN) 

AWS Edge to Power Private Networks and Industry 4.0

At the ONF 2020 Spotlight Series, '5G Connected Edge Cloud for Industry 4.0 Transformation', Ishwar Parulkar, Chief Technologist, Telco, Amazon Web Services (AWS) gave a keynote presentation on 'How AWS Edge is Powering Industry 4.0'. 

He talked about how AWS Edge is powering the 4th Industrial Revolution by building the connected edge cloud. His talks covers:

• Edge Developer Expectations
• AWS Vision of Edge
• Edge Infrastructure
• AWS Enabled Private Networks
• Services for the Connected Edge

The video of the talk is embedded below and the slides are available here.

Back in January, AWS also released a nice whitepaper on 'Next-Generation Mobile Private Networks Powered by AWS'.

This whitepaper introduces the relevant use cases, solutions, and best practices for designing and deploying mobile private networks powered by AWS. Cloud-enabled mobile private networks allow enterprises, governments, and professional organizations to autonomously deploy their own geo-dispersed, secure mobile private networks. These networks cover private facilities while meeting their performance, reliability, availability, security, and scalability requirements. 

It's available to download here.

Related Posts:

• The 3G4G Blog - Webinar: Where Edge Meets Cloud by Dean Bubley
• The 3G4G Blog: 5G Private and Non-Public Network (NPN)
• The 3G4G Blog: What is Industrial IoT (IIoT) and how is it different from IoT?
• The 3G4G Blog: 5G and Industry 4.0
• 3G4G: What is Industrial IoT (IIoT)?
• 3G4G: Industry 4.0, IIoT, IoT & 5G
• The 3G4G Blog: Can Augmented & Mixed Reality be the Killer App 5G needs?
• The 3G4G Blog: What is Cloud Native and How is it Transforming the Networks?
• Telecoms Infrastructure Blog: Edge Computing and the Future of Small Cell Networks
• Free 6G Training - 6G in the sky: On-demand intelligence at the edge of 3D networks 

Edge Computing and the Future of Small Cell Networks

Small Cell Forum (SCF) recently published report setting out how edge computing will impact the future of small cell networks, with particular focus on private 4G / 5G cellular networks. A survey of service providers for the paper showed that by 2025 almost 75% new indoor small cell deployments will be co-located with edge and/or private EPC. The paper (SCF 234) is available to download here.

The press release on this stated:

Edge Computing and Small Cell Networks identifies core synergies between edge computing (EC) and small cell networks (SCNs) and highlights how those synergies are present across multiple domains – business, technical, deployment, product and vendor ecosystems.

It describes how the benefits of small cells co-located with edge can be applied to commercial and operational advantage in key industry segments, such as automation and Industry 4.0, worksites, mission critical services, enterprises and public safety. The paper looks in detail at the architectures and deployment considerations for edge and small cells in three premises-based use cases:

• Fully private cellular networks (PCNs)
• PCNs that have a roaming relationship with MNO networks
• PCNS integrated with MNO networks

While small cells and edge computing have significant potential to meet enterprise demand and drive new business models for service providers, best practice needs to be agreed and adopted, and technical barriers/gaps addressed to optimize that potential for both enterprises and service providers. Key areas of focus include:

• Edge network recommendations: For edge computing solutions, the ‘edge network’ has to work in concert with the ‘core network’ via open interfaces and APIs to enable true multi-vendor ecosystem. SCF believes that current specifications are incomplete and need enhancements.
• Edge platform services and applications recommendations: Open and consistent APIs across multiple organizations must evolve and align to enable a broad ecosystem of edge platform services and edge applications.
• EC platform solution recommendations: Blueprints/reference-designs/solutions for open-source edge computing platforms are urgently required for rapid growth of the EC ecosystem.
• EC and small cell recommendations: Design/deployment blueprints must be available for core use cases to best leverage small cell/edge synergies to deliver multiple benefits: shared virtualized implementations leading to cost efficiencies; integrated network functions & mutually beneficial analytics (radio environment, RAN characteristics, location etc.) leading to advanced functionalities to the edge computing platforms.
• EC infrastructure recommendations: In determining the COTS hardware for optimal edge computing infrastructure, service providers and enterprises shouldn’t be tempted to reinvent the wheel. There are already robust guidelines for data center & hardware design & implementations from TIA, BICSI and OCP that should be followed.

‘The benefits of edge computing are well known, but what we have begun to demonstrate with this work is that small cells and edge will be critical for enabling service providers and enterprises to realize new profitable service opportunities,’ said Dr. Prabhakar Chitrapu, Chair of Small Cell Forum. ‘Small cells plus edge will also enable new business models for a broad spectrum of stakeholders, including edge infrastructure, edge network and edge platform as a service, as well as direct edge application services to subscribers, enterprises and service providers.’

Going forward, working in collaboration with other relevant Industry Forums, SCF will lead the development of a set of harmonized and consistent set of application, network & system-level APIs to enable small cell networks to facilitate EC services and applications.

SCF will also spearhead the deployment of design blueprints for core use cases, leveraging small cell/edge synergies and open source environments.

The research was made possible by an extended collaboration of leading MNOs, OEMs and Infrastructure providers, brought together by Small Cell Forum, including; American Tower, AT&T, Crown Castle, Druid, Intel, Nokia and Reliance Jio.

A video by Small Cell Forum provides a bit more background of this new Edge Computing and Small Cell Networks whitepaper:



Edge computing is a hot topic and every time we have posted something, we have got a lot of engagement on the topic.

The 3G4G Blog - Webinar: Where Edge Meets Cloud by Dean Bubley - https://t.co/TjGnwJ5yh0 #5G #3G4G5G pic.twitter.com/eWa4EQsDtP

— 3G4G (@3g4gUK) May 1, 2019

Edge computing investments: who’s doing what and why? - https://t.co/kZMBRxc6J6 - a nice webinar presented by @REALDaliaAdib, @STLPartners #Free5Gtraining #5G #5GNetworks #URLLC #Latency #MEC #MobileEdge #EdgeCompute #STLPartners #TelecomsInfra #IaaS #PaaS #SaaS pic.twitter.com/jZwv3CWpUS

— 5G Training (@5Gtraining) June 5, 2020

Related Posts:

• Telecoms Infrastructure Blog: SuperMicro's 5G Pole-Mounted DU Server Solution
• The 3G4G Blog: SCF Releases 5G Functional API to Enable Open Small Cells Ecosystem
• The 3G4G Blog - Webinar: Where Edge Meets Cloud by Dean Bubley
• The 3G4G Blog: Can Augmented & Mixed Reality be the Killer App 5G needs?
• The 3G4G Blog: Augmented / Virtual Reality Requirements for 5G
• The 3G4G Blog: 'Mobile Edge Computing' (MEC) or 'Fog Computing' (fogging) and 5G & IoT
• Connectivity Technology Blog: Base Stations, Backhaul & Energy Innovations to Improve Rural Connectivity

Rakuten details how Cloud will make Telecommunications Industry Irrelevant

Yesterday we detailed Rakuten's Open RAN and Innovation Journey here. Today, we look at a presentation from the recent 5G Malaysia International Conference 2020 where Tareq Amin, CTO, Rakuten Mobile, presented his vision on how Cloud will eventually make Telecommunications Industry Irrelevant.

Innovation has allowed them to create a site with the smallest footprint. Not only that, the change of design has allowed the operator to save massive amounts of money and at the same time pushing the software in the cloud  allows the site to come up is 8.5 minutes as compared to the traditional time of between half to two days.

Rakuten uses 6 hardware types in their data centers which are COTS servers and this has helped them achieve some amazing results as can be seen here.

Tareq is expecting Opex to reduce even more due to automation. Time will tell. Here is the presentation embedded below.


Related Posts:

• Operator Watch Blog: Rakuten details its Open RAN and Innovation Journey
• Telecoms Infrastructure Blog: Airspan Small Cells and Macrocells Portfolio, including 5G