Schneider Electric releases its initial integrated rack with immersed, liquid-cooled IT for data centers

Schneider Electric, vendor of digital transformation of energy management and automation, with Avnet and Iceotope, announces creation of commercially-available integrated rack with chassis-based, immersive liquid cooling. Optimized for compute-intensive applications, the solution combines a high-powered GPU server with Iceotope’s liquid cooling technology to increase energy efficiency. 

Avnet integrates the liquid-cooled server with Schneider Electric’s NetShelter liquid-cooled enclosure system for simple deployment into data centers or edge computing environments. The system is EcoStruxure Ready since the solution is available with next generation data center management software, EcoStruxure IT Expert and our digital service EcoStruxure Asset Advisor. 

This liquid-cooled solution is ideal for applications such as big data analytics, artificial intelligence, and machine-learning algorithm training development, where high compute demands more energy use. In a recent report published by Gartner, liquid cooling was identified as a technology to watch. 

Analyst Henrique Cecci advised data center operators “maximize cooling energy efficiencies by employing modern liquid cooling solutions.” Liquid cooling offers greater efficiency, lower operating costs, smaller footprint, increased reliability, and nearly silent operation despite the high-power density of the GPUs. 

Avnet, Iceotope and Schneider Electric plan to expand the offering as demand grows by welcoming other server OEMs into the partnership. 

“Schneider Electric is committed to making data centers more sustainable and liquid cooling is a very compelling approach,” said Kevin Brown, CTO and SVP of Innovation, Secure Power, Schneider Electric, who is presenting on liquid cooling at the Gartner conference. “This latest development marks a significant step toward industrializing chassis-based immersion solutions which offer the efficiency and effectiveness of tanks based solutions while providing the compatibility and serviceability of more traditional, ‘direct-to-chip’ liquid-cooling designs. Given the growth of compute-intensive applications, we believe this approach is very promising.”

“As a leading global technology solutions provider, Avnet has always excelled at adapting to changing markets, trends, and technologies,” said Scott MacDonald, global president, Avnet Integrated. “We do well at this by partnering with the world’s best technology designers and hardware manufacturers. In this case, we’re partnering with liquid cooling-specialist, Iceotope, and global infrastructure giant, Schneider Electric, to make the best possible liquid cooling solutions for our customers. We’re excited to integrate, fulfill, and support this solution that smartly addresses customer cooling challenges around rising chip densities, harsh IT environments, rising energy costs, water use restrictions, and space constraints.”

“Iceotope is thrilled to be making its innovative chassis-level, immersive liquid cooling technology available to the general market,” said David Craig, CEO of Iceotope. “We offer the only liquid cooling tech that scales easily from edge computing devices to large server farms in the cloud. And compared to direct-to-chip liquid cooling systems, our approach is more efficient, more reliable, saves more space, eliminates fans, and when deployed across the data center, it lowers cost.”

Oracle makes it to DISA Impact Level 5 provisional authorization for Oracle Cloud Infrastructure

Following closely on the heels of Oracle achieving FedRAMP authorization, Oracle announces Tuesday three new government regions: Ashburn, Virginia; Phoenix, Arizona; and Chicago, Illinois. These regions have achieved DISA Impact Level 5 provisional authorization (IL5 PATO), providing a cloud environment where U.S. Department of Defense (DoD) and other federal customers can harness the power of Oracle Cloud to unlock innovation, improve mission performance, and enhance service delivery.  

This is an important milestone in Oracle’s journey to deliver innovative cloud services with consistent high performance and exceptional security to the entire U.S. government. In 2020, Oracle plans to bring additional full-scale Gen 2 Cloud Regions online to support the classified missions of the US Government.

“U.S. DoD and other Federal customers are continually looking for new, secure ways to improve citizen services and keep our nation safe,” said Don Johnson, executive vice president, Oracle Cloud Infrastructure. “Oracle’s Generation 2 Cloud was engineered to deliver highly secure, high-performance, cost effective infrastructure that helps government organizations address the needs of the nation today and tomorrow.” 

Oracle has been a key technology partner of the U.S. government. Currently, more than 500 government organizations take advantage of Oracle’s technologies and superior performance. State, local and federal government customers using Oracle to modernize their technology include Defense Manpower Data Center (DMDC) and the U.S. Air Force.

The Department of Defense recently awarded a contract to Oracle for its Oracle Cloud Infrastructure to support a large portion of the enterprise human resource portfolio. The award modernizes existing infrastructure and will assist the Defense Manpower Data Center in providing necessary human resource services and capabilities to its military members, veterans and their families.

With Oracle Cloud Infrastructure, customers benefit from best-in-class security, consistent high performance, simple predictable pricing, and the tools and expertise needed to bring enterprise workloads to cloud quickly and efficiently. In addition, Oracle now provides organizations with a complete set of solutions for any high performance computing (HPC) workload, enabling businesses to capitalize on the benefits of modern cloud computing while enjoying performance comparable to on-premises at a lower cost.

Oracle Cloud Infrastructure has achieved certifications and attestations for key security standards and compliance mandates. These independent third-party assurance programs demonstrate Oracle’s commitment to security and to meeting the needs of the public sector. These IL5 PATO government regions will launch with initial Oracle services including VM and bare metal compute (CPU and GPU), storage (including archive, block, and object storage), database, identity and access management, key management service, load balancer, and Exadata cloud service.

“Oracle Cloud Infrastructure brings incredible performance, flexibility, security, and cost-savings benefits to our federal civilian, commercial and higher education customers,” said Paul Seifert, federal sector president, Mythics. “Mythics’ DoD customers will now be able to leverage Oracle Cloud to better serve the unique requirements of the DoD at home and abroad.”

“We’re excited to see the Oracle Cloud Infrastructure achieve DISA Impact Level 5 provisional authorization, as it provides additional options that our federal government clients—especially those seeking to migrate large and complex Oracle-based solutions to the cloud—can leverage,” said Anthony Flake, managing director of Accenture Federal Services’ Oracle practice.

Schneider Electric brings out its initial integrated rack with immersed, liquid-cooled IT for data centers

Schneider Electric, alongwith Avnet and Iceotope, announced on Monday  creation its commercially-available integrated rack with chassis-based, immersive liquid cooling. Optimized for compute-intensive applications, the solution combines a high-powered GPU server with Iceotope's liquid cooling technology to increase energy efficiency. 

Avnet integrates the liquid-cooled server with Schneider Electric's NetShelter liquid-cooled enclosure system for simple deployment into data centers or edge computing environments. The system is EcoStruxure Ready since the solution is available with next generation data center management software, EcoStruxure IT Expert, and digital service EcoStruxure Asset Advisor. 

This liquid-cooled solution is ideal for applications such as big data analytics, artificial intelligence, and machine-learning algorithm training development, where high compute demands more energy use. In a recent report published by Gartner, liquid cooling was identified as a technology to watch. 

The parallel-processing power of Graphical Processing Units, known as GPUs, makes them efficient processor for a growing number of applications including AI, big data analytics, data mining, and more. The chips are increasingly power dense with thermal design power ratings reaching 400 watts or more. 

This makes traditional data center air-cooled architectures impractical, or costly and less efficient than liquid-cooled approaches where the server is partially submerged in a dielectric fluid. 

Liquid cooling delivers efficiency removing the need for air conditioning, while offering the same processing power with less energy. It is also silent as it does not include industrial drone from fans and pumps; delivers resilience as all components are in a sealed module that resists dust and smoke; and comes in a compact size to offer smaller and more flexible footprint.

Analyst Henrique Cecci advised data center operators "maximize cooling energy efficiencies by employing modern liquid cooling solutions1." Liquid cooling offers greater efficiency, lower operating costs, smaller footprint, increased reliability, and nearly silent operation despite the high-power density of the GPUs. Avnet, Iceotope, and Schneider Electric plan to expand the offering as demand grows by welcoming other server OEMs into the partnership. 

"Schneider Electric is committed to making data centers more sustainable and liquid cooling is a very compelling approach," said Kevin Brown, CTO and SVP of Innovation, Secure Power, Schneider Electric, who is presenting on liquid cooling at the Gartner conference. "This latest development marks a significant step toward industrializing chassis-based immersion solutions which offer the efficiency and effectiveness of tanks based solutions while providing the compatibility and serviceability of more traditional, 'direct-to-chip' liquid-cooling designs.  Given the growth of compute-intensive applications, we believe this approach is very promising."

Diamanti joins AI/ML sector with its GPU platform that supports containerized workloads on Kubernetes

Diamanti announced availability of its enterprise platform with GPU support for running containerized workloads under Kubernetes, ideal for the demanding requirements of emerging artificial intelligence (AI) and machine learning (ML) applications. 

In conjunction with its recent announcement of Diamanti Spektra, customers can now provide to their end users GPU capacity in cloud clusters for scaling AI/ML workloads on premises out to public clouds to accelerate model development and training.

Diamanti recently announced the close of a $35 million Series C funding round that the company plans to use to ramp global go-to-market initiatives, along with increased investment in engineering resources to drive the roadmap for Diamanti Spektra, as well as new software, SaaS, and hardware solutions for emerging AI and ML workloads.

“AI is quickly proving to be the most disruptive set of new technologies in decades thanks to breathtaking advances in computing power, volume, velocity and variety of data,” said Tom Barton, CEO of Diamanti. “Our platform offers unmatched extensibility and flexibility for containerized workloads and now our customers can add GPU support for a heterogeneous environment under the same Kubernetes umbrella to help with even the most demanding AI/ML requirements.”

The Diamanti platform for AI/ML workloads fully supports Nvidia’s NVLink cross connect GPU card technology for higher performing workloads, as well as Kubeflow, a machine learning framework for Kubernetes that provides highly-available Jupyter notebooks and ML pipelines.

“Cloud-native methodology and software are crossing over with AI and machine learning, with Kubernetes an increasingly attractive option for data scientists to orchestrate the distributed architecture required to run multiple machine learning libraries and frameworks in production at scale,” said Matt Aslett, research vice president, 451 Research. “One key use case across several verticals involves infrastructure that can be quickly spun up or down to support massive simulations.”

Early access Diamanti customers are already benefiting from the new platform support for GPUs in industries as varied as financial services, energy and travel, among others.

For AI/ML applications requiring GPUs, the new Diamanti Spektra solution can also now manage the full lifecycle of containerized workloads across on-premises and public clouds, moving applications and data between Kubernetes clusters as necessary. 

Diamanti Spektra combines the power of Diamanti’s hardware-boosted x86 platform along with cloud-based infrastructure to provide Kubernetes-as-a-Service. Diamanti Spektra is in technology preview.

Supermicro debuts server class edge systems for open 5G radio access network (RAN) offerings

Supermicro Computer announced new solutions for 5G cell tower deployments leveraging fully-configurable SuperServers based on second generation Intel Xeon Scalable processors, O-RAN compliant partner software, and ability to operate in harsh environments. These capabilities accelerate the mobile network evolution from proprietary hardware/software to open source software and disaggregated hardware for 5G installations. 

Supermicro's two new systems are its first servers for 5G, the Intelligent Edge, and other embedded applications to be based on second generation Intel Xeon processors. 

The E403-9P-FN2T is built for demanding environments and includes three PCI-E slots for GPU and FPGA accelerator cards. The compact 1U 1019P-FHN2T is ideal for controlled environments such as micro data centers and re-purposed central office locations and features two full-height full-length PCI-E slots.

With these expansion slots, Supermicro can provide real-time Edge AI inferencing via GPU cards, and accelerate 5G RAN software and open-standard site-to-site communication using the Intel FPGA Programmable Acceleration Card N3000. These new servers complement Supermicro's Xeon D-based 1019D and E403 models. 

Supermicro is developing IP65-rated protective enclosures to meet the needs of outdoor environments such as cell towers and microcell sites.

Supermicro is a market leader in delivering high-performance, power-efficient open hardware server platforms extending the data center level remote management to edge computing platforms, and has deep experience with virtualization and containers, including software such as Kubernetes and VMware. 

The 5G architecture is based on open hardware platforms connected by standardized interfaces, and virtualized network elements. Supermicro joined the O-RAN Alliance to help build a cloud-native, open 5G RAN architecture, and these solutions will concurrently support 4G as networks transition to 5G.

The 5G core network has moved to a virtualized model, which runs on data center hardware where Supermicro's Resource-Saving Architecture provides computing platforms such as the high-performance multi-node BigTwin and the high-density SuperBlade and MicroBlade products. These systems leverage shared cooling and efficient power, and feature a disaggregated structure to enable the latest upgrades without the wasteful rip-and-replace of entire server racks.

Amazon SageMaker Operators for Kubernetes capability helps developers, data scientists to train, tune, deploy ML models

AWS released on Monday Amazon SageMaker Operators for Kubernetes capability that makes it easier for developers and data scientists using Kubernetes to train, tune, and deploy machine learning (ML) models in Amazon SageMaker. Customers can install these Amazon SageMaker Operators on their Kubernetes cluster to create Amazon SageMaker jobs natively using the Kubernetes API and command-line Kubernetes tools such as ‘kubectl’.

Many AWS customers use Kubernetes, an open-source general-purpose container orchestration system, to deploy and manage containerized applications, often via a managed service such as Amazon Elastic Kubernetes Service (EKS). This enables data scientists and developers, for example, to set up repeatable ML pipelines and maintain greater control over their training and inference workloads. 

Amazon SageMaker brings down deep learning inference costs by up to 75 percent using Amazon Elastic Inference to attach elastic GPU acceleration to Amazon SageMaker instances. For most models, a full GPU instance is over-sized for inference. Also, it can be difficult to optimize the GPU, CPU, and memory needs of your deep learning application with a single instance type. 

Elastic Inference allows users to choose the instance type that is best suited to the overall CPU and memory needs of your application, and then separately configure the right amount of GPU acceleration required for inference.

However, to support ML workloads these customers still need to write custom code to optimize the underlying ML infrastructure, ensure high availability and reliability, provide data science productivity tools, and comply with appropriate security and regulatory requirements. 

For example, when Kubernetes customers use GPUs for training and inference, they often need to change how Kubernetes schedules and scales GPU workloads in order to increase utilization, throughput, and availability. Similarly, for deploying trained models to production for inference, Kubernetes customers have to spend additional time in setting up and optimizing their auto-scaling clusters across multiple Availability Zones.

Amazon SageMaker Operators for Kubernetes bridges this gap, and customers are now spared all the heavy lifting of integrating their Amazon SageMaker and Kubernetes workflows. Starting Monday, customers using Kubernetes can make a simple call to Amazon SageMaker, a modular and fully-managed service that makes it easier to build, train, and deploy machine learning (ML) models at scale. 

With workflows in Amazon SageMaker, compute resources are pre-configured and optimized, only provisioned when requested, scaled as needed, and shut down automatically when jobs complete, offering near full utilization. 

Now with Amazon SageMaker Operators for Kubernetes, customers can continue to enjoy the portability and standardization benefits of Kubernetes and EKS, along with integrating the many additional benefits that come out-of-the-box with Amazon SageMaker, no custom code required.

Each Amazon SageMaker Operator for Kubernetes provides users with a native Kubernetes experience for creating and interacting with jobs, either with the Kubernetes API or with Kubernetes command-line utilities such as kubectl. Engineering teams can build automation, tooling, and custom interfaces for data scientists in Kubernetes by using these operators—all without building, maintaining, or optimizing ML infrastructure. 

Data scientists and developers familiar with Kubernetes can compose and interact with Amazon SageMaker training, tuning, and inference jobs natively, as users would with Kubernetes jobs executing locally. Logs from Amazon SageMaker jobs stream back to Kubernetes, allowing consumers to natively view logs for model training, tuning, and prediction jobs in command line.

Intel releases tiny EMIB that assist chips to ‘talk’ with each other

Intel unveiled its Embedded Multi-die Interconnect Bridge (EMIBs) chips that speed the flow of data inside nearly 1 million laptops and field programmable gate array devices worldwide. That number will soon soar and include more products as EMIB technology enters the mainstream. 

EMIB is a cost-effective approach to in-package high density interconnect of heterogeneous chips. It delivers a complex multi-layered sliver of silicon no bigger than a grain of rice, and lets chips fling enormous quantities of data back and forth among adjoining chips at blinding speeds: several gigabytes per second.

The industry refers to this application as 2.5D package integration. Instead of using a large silicon interposer typically found in other 2.5D approaches, EMIB uses a very small bridge die, with multiple routing layers. This bridge die is embedded as part of the substrate fabrication process.

For example, Intel’s Ponte Vecchio processor, a general-purpose GPU the company unveiled this month, contains EMIB silicon.

To meet customers’ unique needs, this innovative technology allows chip architects to cobble together specialized chips faster than ever. And compared with an older, competing design called an interposer — in which chips inside a package sit atop what is essentially a single electronic baseboard, with each chip plugged into it — tiny, flexible, cost-effective EMIB silicon offers an 85 percent increase in bandwidth. That can make tech — laptop, server, 5G processor, graphics card— run faster, and next-generation EMIB could double or even triple that bandwidth.

The silicon interposer in a typical 2.5D package is a piece of silicon larger-than-all interconnecting die. In contrast, the silicon bridge is a small piece of silicon embedded only under the edges of two interconnecting die. This allows for most size die to be attached in multiple dimensions, eliminating additional physical constraints on heterogeneous die attachment within the theoretical limits.

The image shows a difficult yet desirable layout. The industry standard 2.5D solution cannot accommodate this, as the silicon interposer cannot be produced large enough to connect all the die. Yet EMIB allows for the flexibility in this die placement, allowing scaling in both dimensions.