Monday, January 10, 2022

AWS adds EC2 instances for HPC workloads powered by AMD

AWS introduced new Amazon Elastic Compute Cloud (Amazon EC2) Hpc6a instances, powered by 3rd Gen AMD EPYC processors, for tightly coupled high performance computing (HPC) workloads. 

Hpc6a instances promise up to 65% better price performance compared to similar compute-optimized Amazon EC2 instances for HPC workloads and can be scaled to carry out complex calculations across a range of cluster sizes—up to tens of thousands of cores. 

Hpc6a instances are enabled with Elastic Fabric Adapter (EFA)—a network interface for Amazon EC2 instances—by default. With EFA networking, customers benefit from low latency, low jitter, and up to 100 Gbps of EFA networking bandwidth to increase operational efficiency and drive faster time-to-results for workloads that rely on inter-instance communications. Hpc6a instances are powered by 3rd Gen AMD EPYC processors that run at frequencies up to 3.6 GHz and provide 384 GB of memory. Using Hpc6a instances, customers can more cost-effectively tackle their biggest and most difficult academic, scientific, and business problems with HPC, and realize the benefits of AWS with superior price performance.

“By consistently innovating and creating new purpose-built Amazon EC2 instances for virtually every type of workload, AWS customers have realized huge price performance benefits for some of today’s most business-critical applications. While high performance computing has helped solve some of the most difficult problems in science, engineering, and business, effectively running HPC workloads can be cost-prohibitive for many organizations,” said David Brown, Vice President of Amazon EC2 at AWS. “Purpose-built for HPC workloads, Hpc6a instances now help customers realize up to 65% better price performance for their HPC clusters at virtually any scale, so they can focus on solving the biggest problems that matter to them most without the cost barriers that exist today.”

“We are excited to continue our momentum with AWS and provide their customers with this new, powerful instance for high performance computing workloads,” said Dan McNamara, Senior Vice President and General Manager, Server Business at AMD. “AMD EPYC processors are helping customers of all sizes solve some of their biggest and most complex problems. From small universities to enterprises to large research facilities, Hpc6a instances powered by 3rd Gen AMD EPYC processors open up the world of powerful HPC performance with cloud scalability to more customers around the world.”

http://www.aws.amazon.com/ec2/instance-types/hpc6

Windstream Wholesale lights 2 Tbps of LH capacity in under 5 days

Windstream Wholesale turned up approximately 2 Terabits of long-haul wave capacity in less than five business days for an unnamed communications provider.

Windstream Wholesale said its ICON network enabled the custom design to be provided within a day. The customer challenged the Windstream Wholesale team to deliver all the bandwidth within 14 days. As its continued success has demonstrated, the fast and flexible team was up for the challenge and exceeded the customer’s expectations — the bandwidth was delivered in less than five days.

“We believe in strong partnerships that step up for each other in times of need. We were delighted to do just that for a company that had a vital bandwidth need,” said Joe Scattareggia, executive vice president of Windstream Wholesale. “Windstream Wholesale’s ‘fast and flexible’ culture to design, cost, quote, sell, and deliver 100 Gig Waves from Chicago to Seattle in a compressed timeframe is something none of our competitors could match. In fact, our services were delivered before the customer ever received a quote from some of their other providers. Windstream Wholesale continues to be a leader and a difference maker in the industry, another great example of why others are partnering with our award-winning team.”

https://www.windstreamenterprise.com/wholesale/interactive-map

Beijing's Capital Online picks Nokia's 7750 Service Router

Capital Online, a cloud computing service provider headquartered in Beijing, will deploy the Nokia 7750 Service Router (SR) and 7250 IXR interconnect router platforms to upgrade its IP backbone network.

The network upgrade entails that the converged backbone provides performance certainty for all traffic flows under all network conditions, the versatility to converge edge and core routing functions onto a common platform, smart traffic engineering with segment routing-MPLS, and granular QoS to address different traffic demands for reliable service delivery.

Xu Xiaohu, Chief Architect of Capital Online, said: “As a trusted partner of critical networks, Nokia has abundant experiences helping its global customers build high capacity and quality IP networks. We’re looking forward to collaborating with Nokia to accelerate the transformation of our network to provide faster, more reliable network services for our global customers throughout the U.S., Europe and Asia.”

Markus Borchert, CEO of Nokia Shanghai Bell, said: “Capital Online is known for being specialists in delivering exceptional online experiences for global business. We are pleased to help it create an industry leading IP network that will upgrade its existing backbone network to the Network 2.0 era, paving the way to increase its footprint in the global Cloud market in the future.”

KT, NTT DOCOMO and Fujitsu test O-RAN in Korea

The KT Research and Development Center in Seoul, South Korea, has tested Fujitsu’s Open RAN based 5G base station equipment in trials for open fronthaul. NTT DOCOMO provided Fujitsu with technical support throughout the project. With the construction of this new test facility, KT will accelerate the introduction of Open RAN technology to Korea’s 5G network.

The three companies have concluded a Memorandum of Understanding centered on the introduction of software-defined virtualized RAN and RAN Intelligent Controllers (RICs).

The three companies agreed to cooperate towards further activities including the construction of an O-RAN test facility and multi-vendor interoperability testing in Korea.

KT is considering the introduction of multi-vendor Open RAN to reduce equipment procurement and construction costs and to achieve flexible network construction capabilities. To this end, KT and Fujitsu constructed an Open RAN verification facility and conducted testing for multi-vendor interoperability with O-RAN open fronthaul in October 2021. 

https://www.fujitsu.com/global/about/resources/news/press-releases/2022/0106-01.html

Norway reports cut on Svalbard cable

Space Norway AS, which owns and operates a subsea cable system connecting mainland Norway with the archipelago of Svalbard, reported a fiber cut on Friday 7 January 2022. The cut is believed to have occurred between 130 and 230 km from Longyearbyen and in an area where the cable goes steeply into the deep sea.  A second cable is still in operation, but the system will have no further redundancy until a repair is completed. A cable-laying ship will be mobilized.

Svalbard is located about midway (74° to 81° north latitude) between the northern coast of Norway and the North Pole. The islands have a population of about 2,900 and the largest settlement is the town of Longyearbyen.

https://spacenorway.no

Atlantic Broadband rebrands as Breezeline

Atlantic Broadband, the eighth-largest cable operator in the U.S., has rebranded as Breezeline.

The rebranding follows the company’s acquisition of two cable systems in Cleveland and Columbus, Ohio in September 2021, which expanded the company’s serviceable households and businesses to more than 1.6 million. 

“We’re no longer just an east coast provider, and we’ve long offered much more than broadband, so our company identity must evolve with us,” said Frank van der Post, President of Breezeline. “The name Breezeline marks the beginning of a new, exciting era of transforming our company through new growth, while also elevating the customer experience through enhanced customer care options, innovative products, and investment in the latest technologies.”

In addition to recent growth through acquisitions, the company has launched a major fiber expansion initiative that will extend connectivity to more than 70,000 additional homes and businesses in New Hampshire and West Virginia via ultra-fast Fiber-to-the-Home (FTTH) technology.



Intel names Michelle Johnston Holthaus as GM of Client Computing

Intel has appointed Executive Vice President (EVP) Michelle Johnston Holthaus to lead the company’s Client Computing Group (CCG). She will be responsible for all aspects of running and growing the client business, including strategy, financial performance and product development for the full portfolio of client technologies and platforms designed to enable exceptional personal computing experiences.

A 25-year Intel veteran, Holthaus brings a deep understanding of Intel’s customers and the client computing business based on her current role as EVP and general manager of the Sales, Marketing and Communications Group, a role she has held since 2017, and her previous experience as the head of global client computing sales. Earlier roles in Holthaus’ tenure include leadership of the Microsoft global account team; management of channel products; central marketing and operations for the PC client group; and management of the reseller product group. Holthaus joined Intel in 1996.

“Michelle’s track record of success driving global sales and revenue for the last five years, combined with her profound understanding of the client computing business and trusted relationships across the entire industry, make her a natural choice to lead our largest business,” said Pat Gelsinger, Intel CEO. “Michelle is a proven leader who embodies Intel’s values, and I look forward to partnering with her in this new capacity as we drive innovation and unquestioned product leadership across the client business.”

Holthaus replaces EVP Gregory Bryant (“GB”), who will leave the company at the end of January for a new opportunity. Since joining Intel in 1992, Bryant has held numerous roles, including most recently as general manager of the Client Computing Group and previously as the general manager of Intel’s Asia Pacific and Japan region.