{"data":{"news":{"items":[{"title":"Keysight Introduces AI Data Center Builder to Validate and Optimise Network Architecture and Host Design","date":"2025-04-28 14:57:00","id":5819,"author":"Eric Ooi","blurb":"","blurb_image_uid":"https://dxfpracgmlczp.cloudfront.net/2025/04/1745823602480_Keysight-KAI-Data-Center-Builder_1600x1600.jpg","blurb_expanded_image_uid":null,"article":"\n\t\n\n\n

Keysight Technologies, Inc. has introduced the Keysight AI (KAI) Data Center Builder — an advanced software suite that emulates real-world workloads to assess the impact of new algorithms, components, and protocols on AI training performance. The KAI Data Center Builder integrates large language model (LLM) and other AI training workloads into the design and validation of key infrastructure components, including networks, hosts, and accelerators. By enabling workload emulation early in the design process, this solution fosters tighter integration between hardware design, protocols, architectures, and AI training algorithms, ultimately enhancing overall system performance.

\n\n

AI operators use various parallel processing strategies, also known as model partitioning, to accelerate AI model training. Aligning model partitioning with AI cluster topology and configuration enhances training performance. During the AI cluster design phase, critical questions are best answered through experimentation. Many of the questions focus on data movement efficiency between the graphics processing units (GPUs). Key considerations include:

\n\n

Scale-up design of GPU interconnects inside an AI host or rack
Scale-out network design, including bandwidth per GPU and topology
Configuration of network load balancing and congestion control
Tuning of the training framework parameters

\n\n

The KAI Data Centre Builder workload emulation solution reproduces network communication patterns of real-world AI training jobs to accelerate experimentation, reduce the learning curve necessary for proficiency, and provide deeper insights into the cause of performance degradation, which is challenging to achieve through real AI training jobs alone. Keysight customers can access a library of LLM workloads like GPT and Llama, with a selection of popular model partitioning schemas like Data Parallel (DP), Fully Sharded Data Parallel (FSDP), and three-dimensional (3D) parallelism.

\n\n

Using the workload emulation application in the KAI Data Center Builder enables AI operators to:

\n\n

Experiment with parallelism parameters, including partition sizes and their distribution over the available AI infrastructure (scheduling)
Understand the impact of communications within and among partitions on overall job completion time (JCT)
Identify low-performing collective operations and drill down to identify bottlenecks
Analyse network utilisation, tail latency, and congestion to understand the impact they have on JCT

\n\n

The KAI Data Centre Builder’s new workload emulation capabilities enable AI operators, GPU cloud providers, and infrastructure vendors to bring realistic AI workloads into their lab setups to validate the evolving designs of AI clusters and new components. They can also experiment to fine-tune model partitioning schemas, parameters, and algorithms to optimise the infrastructure and improve AI workload performance.

\n\n

Ram Periakaruppan, Vice President and General Manager, Network Test & Security Solutions, Keysight, said: \"As AI infrastructure grows in scale and complexity, the need for full-stack validation and optimization becomes crucial. To avoid costly delays and rework, it’s essential to shift validation to earlier phases of the design and manufacturing cycle. KAI Data Centre Builder’s workload emulation brings a new level of realism to AI component and system design, optimizing workloads for peak performance.”

\n\n

KAI Data Centre Builder is the foundation of the Keysight Artificial Intelligence (KAI) architecture, a portfolio of end-to-end solutions designed to help customers scale artificial intelligence processing capacity in data centres by validating AI cluster components using real-world AI workload emulation.

\n\n

\n\n\n","article_image_uid":"https://dxfpracgmlczp.cloudfront.net/2025/04/1745823605106_Keysight-KAI-Data-Center-Builder_1600x1600.jpg","article_expanded_image_uid":null,"article_image_caption":"","publications":"5","tags":"data centre,Keysight Technologies","category":"Products & Technologies"},{"title":"SentinelOne launches Singapore data centre","date":"2022-08-07 22:37:18","id":3978,"author":"Ngai Meng Chan","blurb":"","blurb_image_uid":"https://dxfpracgmlczp.cloudfront.net/2022/08/1659883105429_40944301951_cc5762c413_c.jpg","blurb_expanded_image_uid":null,"article":"\n\t\n\n\n

SentinelOne, an autonomous cybersecurity platform company, has announced the launch of its new data centre in Singapore. The deployment of the new data centre will aid Singaporean businesses and government agencies in their overall data protection and regulatory compliance efforts.
\n
\nHighly regulated industries, including financial services providers, government agencies, and critical infrastructure companies, protect customer data by storing the information on servers located in Singapore. SentinelOne’s data centre deployment provides a path to the cloud for these industries, allowing them to benefit locally from the protection of SentinelOne’s market-leading cybersecurity.
\n
\nLeveraging the cloud is a critical factor for organisations looking to improve their security posture. From a product perspective, cloud solutions are faster and easier to deploy and manage. \"The Singapore data centre is a big step toward upgrading the security posture of Singaporean enterprises while helping to achieve compliance with data protection laws,\" said Raj Rajamani, Chief Product Officer at SentinelOne. \"The cloud enables best-in-class protection and unprecedented scale, enabling customers to quickly deploy and standardise on Singularity XDR.\"
\n
\nSentinelOne said the data centre comes at an essential time for Singapore. In 2021, over 65% of Singaporean organisations were hit with ransomware, up from just 25% in 2020. The ransom has risen as well. Organisations that had their data encrypted by attackers paid an average of US$1.16M million in 2021, six times higher than the $187,500 average in 2020. These increases demonstrate the need Singaporean organisations have for modernising their cybersecurity.
\n
\nThe data centre, built on the AWS platform, protects against cyberattacks with investments in preventing impact from power failures, natural disasters, and break-ins beyond those of most individual organisations.

\n\n\n","article_image_uid":"https://dxfpracgmlczp.cloudfront.net/2022/08/1659883123886_40944301951_cc5762c413_c.jpg","article_expanded_image_uid":null,"article_image_caption":"","publications":"5","tags":"Singapore,data centre,SentinelOne,data center","category":"Projects"},{"title":"Core Data Centres, iLOQ and Experienced System Integrator Works Together for Maximised Data Center Security ","date":"2021-11-19 08:45:10","id":3335,"author":"Ming En Liew","blurb":"","blurb_image_uid":"https://dxfpracgmlczp.cloudfront.net/2021/11/1637246868722_iLoq.png","blurb_expanded_image_uid":null,"article":"\n\t\n\n\n

A new locking system that promises smooth and secure access to servers in the Canadian data centre has been formulated by data centre operator Core Data Centres Inc., digital and mobile access firm iLOQ and an experienced system integrator.

\n\n

For years, Core Data Centres has owned and operated state-of-the-art data centers in Canada. With the company providing solutions ranging from cloud and information technology services to managed services and colocation, maintaining high levels of security is vital. With their numerous customers, there is a constant flow of people needing to access their servers, increasing the risk of threats.

\n\n

Breaches in security by unauthorised individuals can have severe consequences for both the company managing the center and for the blue-chip clients whose critical data has been penetrated. A secure environment is thus crucial for the continued success of Core Data Centres.

\n\n

Yet, traditional access management systems on the market come with issues such as a high price point and space constraints, as there is a need for power source and cabling.

\n\n

Seeking a solution, Core Data Centres worked together with an experienced system integrator and iLOQ to integrate the iLOQ S50 mobile access sharing solution into their data centres.

\n\n

With iLOQ S50, a smartphone acts as both a key and a power source. The cylinders are NFC powered and can be installed on the swing handles of the racks in a matter of minutes. No batteries or cabling are needed, reducing costs and environmental impact.

\n\n

Security risks related to physical keys have also been eliminated. Access rights are sent as a mobile key and remotely shared only with the authorised individuals. Features like time-restricted access, instant granting and revoking of access rights, and real-time audit trail reports optimise security. Further, the entire system can be managed with just one software program.

\n\n

iLOQ NFC-powered locking cylinders will provide secure and convenient access to the server racks while iLOQ S50 padlocks will be installed on the standby generators and the gas lines feeding them.

\n\n

“This is a highly secure building, and our operations demand the same level of security. Handing over physical keys in person can be a major security risk. A major bonus of iLOQ S50 is that the security desk can quickly and easily send, and also cancel, mobile keys over the air. We have full control at all times and can be sure only the right people have permission to access the servers,” states Bill Henneberry, Chief Technology Officer, Core Data Centres.

\n\n\n","article_image_uid":"https://dxfpracgmlczp.cloudfront.net/2021/11/1637246870009_iLoq.png","article_expanded_image_uid":null,"article_image_caption":"Photo by: iLoq","publications":"5","tags":"iLoq,core data centres inc","category":"Products & Technologies"}],"pageInfo":{"hasNextPage":false,"hasPreviousPage":false,"totalCount":3,"totalPages":0,"currentIndex":0}}}}