Back to Table of contents

Primeur live 2019-11-21

Exascale

NCI a founding member of Intel-Lenovo Project Everyscale  ...

Focus on Europe

ADVA and PSNC showcase 1200G channel transmission at SC19 ...

200 Gigabit HDR InfiniBand to accelerate Eni supercomputing platform, to become world's most powerful industrial supercomputer ...

IT4Innovations National Supercomputing Center, a EuroHPC supercomputing site, selected Mellanox 200G HDR InfiniBand to accelerate their new research infrastructure ...

Barcelona Supercomputing Center takes one of top spots in the IO500's "10-Node Challenge" at SC19 ...

Middleware

Univa announces advanced workload management for NVIDIA GPUs on Arm-based systems ...

Bright Computing makes award winning Bright Cluster Manager software free for clusters up to 8 nodes ...

NVIDIA launches Magnum IO software suite to help eliminate data bottlenecks for data scientists and AI, HPC researchers ...

Hardware

San Diego Supercomputer Center supersizes advanced storage with Qumulo for its data-intensive infrastructure ...

Liqid and Western Digital collaborate to deliver fully managed NVMe-oF open composable platform for shared accelerated storage ...

Liqid showcases software disaggregation for converged servers ...

TYAN showcases HPC, AI and storage server platforms featuring 2nd Gen Intel Xeon Scalable processors at SC19 ...

Beijing Meteorological Service selects 200 Gigabit HDR InfiniBand to accelerate new supercomputing platform ...

TMGcore unveils groundbreaking data centre platform at Supercomputing 2019 ...

Argonne National Laboratory deploys Cerebras CS-1, the world's fastest Artificial Intelligence computer ...

KAUST installs Cray CS-Storm 500NX supercomputer to drive strategic AI initiative ...

NVIDIA and tech leaders team to build GPU-accelerated Arm servers for new era of diverse HPC architectures ...

Penguin Computing to deliver Magma supercomputer, one of the first Intel Xeon Platinum 9200 Series processor-based servers for AI and HPC ...

Penguin Computing upgrades Corona with latest AMD Radeon Instinct GPU technology for enhanced ML and AI capabilities ...

San Diego Supercomputer Center selects Dell Technologies to power "Expanse" system ...

Supermicro collaborates with Intel to deliver large scale distributed training AI systems ...

Inspur showcases new liquid cooling HPC system with natural circulation evaporative cooling technology at SC19 ...

HDR InfiniBand selected by Meteo France to accelerate two new large scale supercomputers ...

Pavilion to provide NVMe-Over-Fabrics storage for SCinet'19 high performance computing (HPC) simulations ...

Applications

Registration for ASC20 Student Supercomputer Challenge ...

TOP500

CSC's Puhti ranked 232 on the latest supercomputer TOP500 list  ...

NCI supercomputer climbs TOP500 rankings  ...

The Cloud

2019 demand for Rescale-managed Cloud HPC exceeds all previous years combined ...

The Kinetica Active Analytics Platform and RAPIDS now available on Oracle Cloud Infrastructure to accelerate predictive data analytics performance ...

NVIDIA announces scalable GPU-accelerated supercomputer in the Microsoft Azure Cloud ...

Company news

ScaleMP to showcase SMP supremacy and composable computing innovations at SC19 ...

Qumulo demonstrates unified file storage at SC'19 that reduces administrative burden while consolidating HPC workloads ...

Dell Technologies introduces new solutions to advance high performance computing and AI innovation  ...

Supermicro delivers industry-leading portfolio of advanced HPC systems at SC19 ...

WekaIO unveils WIN, a global partner programme ...

Liqid and Western Digital collaborate to deliver fully managed NVMe-oF open composable platform for shared accelerated storage

19 Nov 2019 Denver - Liqid, provider of the world's most comprehensive composable infrastructure platform, has signed a collaboration with Western Digital to demonstrate NVMe over Fabrics (NVMe-oF), leveraging Liqid's software to manage the OpenFlex open composable platform across low-latency Ethernet fabrics. Liqid Command Center 2.2 software also composes the OpenFlex platform in tandem with GPU and CPU resources up and down the hardware rack and across multiple racks for multi-petabyte storage capacity at NVMe speeds. By enabling IT users to aggregate and share large pools of data centre resources over distance, while also allowing for more granular resource management and automation at the stack level, Western Digital and Liqid are showcasing an expansive, multi-fabric composable system designed to dynamically maximize resource utilization and performance wherever physical hardware resides.

"Our OpenFlex composable platform is purpose-built for disaggregated deployment in the world's most data-intensive environments that require agile, optimized storage performance for diverse and unpredictable workflows", stated Scott Hamilton, Senior Director of Product Management in Western Digital's Data Center Business Unit. "Liqid's Command Center software leverages Western Digital's Open Composable API to orchestrate OpenFlex platform resources enabling faster NVMe-oF deployments and further expanding the scope of composable resources. With OpenFlex and Liqid, IT users can create more precisely balanced, disaggregated systems that can scale to meet evolving business requirements."

OpenFlex E3000 and F3100 Series from Western Digital allows storage to be disaggregated from compute, enabling applications to share a common pool of NVMe storage capacity. Data can easily be shared between applications, or needed capacity can be allocated to an application regardless of physical location combining multiple, modular OpenFlex storage fabric devices in a single rack or across multiple enclosures.

OpenFlex E3000 and F3100 series devices feature:

  • Access to data from anywhere in the data center via ultra-low latency NVMe-oF
  • Up to 61,4TB capacity per device, each featuring dual-port 50GbE fabric connections
  • Up to 10 F3100 devices per E3000 enclosure
  • Massive NVMe performance delivering up to 110 GB/s and 22 million IOPS per OpenFlex E3000 enclosure
  • Deliver the data locality benefits of direct-attached NVMe (low latency, high performance) with the agility and flexibility of composable storage
  • Hot-swappable power supplies, fans, and fabric devices ensure maximum availability without operational interruption

Liqid's Command Center 2.2 composable software is fully integrated with Western Digital's Open Composability API to enable manageability and automation of the full appliance enclosure, including data volume management, sensor monitoring, power usage, fan speeds, and other critical activities. Compose for resources inside the appliance, and connect multiple OpenFlex-based racks together with Liqid software, sharing connected devices across the data center, and creating multiple petabytes of dynamic, pooled and managed OpenFlex-driven storage.

"By coupling Liqid's award-winning, multi-fabric software with the powerful OpenFlex platform from Western Digital, we deliver fully composable NVMe-oF, with a disaggregated system built from the ground up for adaptive performance associated with artificial intelligence and other data-intensive applications", stated Marius Tudor, vice president, Sales & Marketing, Liqid. "In addition, Liqid Command Center 2.2 enables IT users to connect the OpenFlex platform with additional disaggregated hardware such as FPGA, NVMe, and Intel Optane memory technology, for applications that can benefit from GPU-oF performance to deliver industry-leading, dynamic, highly scalable data centre environments."

Liqid Command Center delivers unified multi-fabric support for composability across all major fabric types including PCIe Gen 3, PCIe Gen 4, Ethernet, Infiniband, and lays the foundation for the upcoming Gen-Z and CXL specifications. Liqid’s upcoming release of its Command Center 2.2 software will enable IT users to dynamically compose servers from pools of CPU, GPU, FPGA, NVMe, Intel Optane memory technology, and networking, regardless of underlying fabric type.

Source: Liqid

Back to Table of contents

Primeur live 2019-11-21

Exascale

NCI a founding member of Intel-Lenovo Project Everyscale  ...

Focus on Europe

ADVA and PSNC showcase 1200G channel transmission at SC19 ...

200 Gigabit HDR InfiniBand to accelerate Eni supercomputing platform, to become world's most powerful industrial supercomputer ...

IT4Innovations National Supercomputing Center, a EuroHPC supercomputing site, selected Mellanox 200G HDR InfiniBand to accelerate their new research infrastructure ...

Barcelona Supercomputing Center takes one of top spots in the IO500's "10-Node Challenge" at SC19 ...

Middleware

Univa announces advanced workload management for NVIDIA GPUs on Arm-based systems ...

Bright Computing makes award winning Bright Cluster Manager software free for clusters up to 8 nodes ...

NVIDIA launches Magnum IO software suite to help eliminate data bottlenecks for data scientists and AI, HPC researchers ...

Hardware

San Diego Supercomputer Center supersizes advanced storage with Qumulo for its data-intensive infrastructure ...

Liqid and Western Digital collaborate to deliver fully managed NVMe-oF open composable platform for shared accelerated storage ...

Liqid showcases software disaggregation for converged servers ...

TYAN showcases HPC, AI and storage server platforms featuring 2nd Gen Intel Xeon Scalable processors at SC19 ...

Beijing Meteorological Service selects 200 Gigabit HDR InfiniBand to accelerate new supercomputing platform ...

TMGcore unveils groundbreaking data centre platform at Supercomputing 2019 ...

Argonne National Laboratory deploys Cerebras CS-1, the world's fastest Artificial Intelligence computer ...

KAUST installs Cray CS-Storm 500NX supercomputer to drive strategic AI initiative ...

NVIDIA and tech leaders team to build GPU-accelerated Arm servers for new era of diverse HPC architectures ...

Penguin Computing to deliver Magma supercomputer, one of the first Intel Xeon Platinum 9200 Series processor-based servers for AI and HPC ...

Penguin Computing upgrades Corona with latest AMD Radeon Instinct GPU technology for enhanced ML and AI capabilities ...

San Diego Supercomputer Center selects Dell Technologies to power "Expanse" system ...

Supermicro collaborates with Intel to deliver large scale distributed training AI systems ...

Inspur showcases new liquid cooling HPC system with natural circulation evaporative cooling technology at SC19 ...

HDR InfiniBand selected by Meteo France to accelerate two new large scale supercomputers ...

Pavilion to provide NVMe-Over-Fabrics storage for SCinet'19 high performance computing (HPC) simulations ...

Applications

Registration for ASC20 Student Supercomputer Challenge ...

TOP500

CSC's Puhti ranked 232 on the latest supercomputer TOP500 list  ...

NCI supercomputer climbs TOP500 rankings  ...

The Cloud

2019 demand for Rescale-managed Cloud HPC exceeds all previous years combined ...

The Kinetica Active Analytics Platform and RAPIDS now available on Oracle Cloud Infrastructure to accelerate predictive data analytics performance ...

NVIDIA announces scalable GPU-accelerated supercomputer in the Microsoft Azure Cloud ...

Company news

ScaleMP to showcase SMP supremacy and composable computing innovations at SC19 ...

Qumulo demonstrates unified file storage at SC'19 that reduces administrative burden while consolidating HPC workloads ...

Dell Technologies introduces new solutions to advance high performance computing and AI innovation  ...

Supermicro delivers industry-leading portfolio of advanced HPC systems at SC19 ...

WekaIO unveils WIN, a global partner programme ...