Data center bridging

Broadcom Announces Industry-Leading Connectivity Portfolio for Hyperscale Video Storage at Scale

Retrieved on: 
Thursday, October 13, 2022
Cloud, SAN, CA, AVGO, Data center bridging, Growth, FC-HBA API, ROCE, Family, NASDAQ, Silicon, IO, HBA, RDMA, HDD, Central processing unit, CPU, Energy, Automation, VP9, Software, Socs, 9600, Human magnetism, SOC, NIC, Ecosystem, IP, OCP, Social media, GLOBE, Tomahawk, AV1, TBPS, Connected, Solution, HAMR, SAS, Video game console, Computer data storage, Internet service provider, Broadcom Inc.

SAN JOSE, Calif., Oct. 13, 2022 (GLOBE NEWSWIRE) -- Broadcom Inc. (NASDAQ:AVGO), the market leader in hyperscale video storage connectivity, today announced it is expanding its silicon, software and hardware storage connectivity products with enhanced performance, capabilities and power.

Key Points: 
  • SAN JOSE, Calif., Oct. 13, 2022 (GLOBE NEWSWIRE) -- Broadcom Inc. (NASDAQ:AVGO), the market leader in hyperscale video storage connectivity, today announced it is expanding its silicon, software and hardware storage connectivity products with enhanced performance, capabilities and power.
  • This expanded product portfolio is the result of multi-generational investments that are intended to aid hyperscale data centers to manage and maintain the storage connectivity they need to deliver video storage at scale.
  • Our leading-edge video processing solutions are setting new benchmarks in video storage delivering the lowest power and latency, while delivering the highest performance in the industry all Connected by Broadcom.
  • The portfolio connects components inside the storage network, to the storage systems and ultimately, storage media.

Broadcom and Arista Deliver Industry-First Open, Performance-Optimized, End-to-End RDMA Over Converged Ethernet (RoCE) Solution

Retrieved on: 
Tuesday, October 11, 2022
SAN, CA, AVGO, Data center bridging, Network interface controller, Meta, Machine learning, ROCE, Solution, Broadcom Inc., NASDAQ, RDMA, Ethernet, Â, Technology, Automation, HPC, Industry, AI, Ecosystem, OCP, NIC, GLOBE, Tomahawk, Remote direct memory access, Documentation, Internet service provider, Computer data storage, Arista

Hyperscaler and enterprise data center operators can now deploy fully optimized systems, connected by Broadcom, using open-standard RDMA.

Key Points: 
  • Hyperscaler and enterprise data center operators can now deploy fully optimized systems, connected by Broadcom, using open-standard RDMA.
  • Jointly integrated by Ethernet market leaders Broadcom and Arista, this end-to-end solution features Broadcoms Ethernet NIC and switch technology.
  • The Broadcom and Arista RoCE solution simplifies the deployment of RoCE in the data center, by utilizing an end-to-end open congestion control algorithm to deliver the industrys lowest latency at scale.
  • Broadcom and Arista are committed to ongoing innovation and will continue to add more products to this solution in the future.

Kalray’s Flashbox™ Wins Coveted Flash Memory Summit Award for Most Innovative Technology

Retrieved on: 
Thursday, August 4, 2022
Data Management, Semiconductor, Technology, Other Technology, Telecommunications, Internet, Hardware, Cloud, RAID, Solid-state drive, IOPS, AI, RDMA, Canada Fitness Award Program, U.2, FMS, Architecture, Software, Data-rate units, NVM Express, MPPA, Multimedia, Data center bridging, Â, NXP Semiconductors, Technology, Santa Clara Convention Center, SDS, Edge computing, CEA, JBOD, Solution, DPU, Central processing unit, OEM, Euronext Growth, Renault–Nissan–Mitsubishi Alliance, PCI Express, IOP, View, Acceleration, Video game console, Computer data storage, Flash, Sanmina Corporation, Kalray

The Flash Memory Summit Awards represent an industry recognition of innovation that Kalrays Flashbox solution brings to the market.

Key Points: 
  • The Flash Memory Summit Awards represent an industry recognition of innovation that Kalrays Flashbox solution brings to the market.
  • Flashbox is a new concept of All-Flash-Array, based on Kalrays MPPA DPU processor (Data Processing Unit) and co-designed with Viking Enterprise Solutions.
  • It is a great honor to win a Flash Memory Summit award for our innovative concept of Flashbox All-Flash Array.
  • Flashbox is a unique high-performance, low power, cost-effective, high-available new generation of storage appliances based on Kalrays DPU technology.

InfiniBand Continues Leadership on TOP500 with 17 Percent Growth; 39th InfiniBand and RoCE Plugfest Bring Record Turnout

Retrieved on: 
Wednesday, July 6, 2022
Mobile, Wireless, Technology, 5G, Carriers and Services, Telecommunications, Software, Networks, Internet, Hardware, Data Management, SFP, Efficiency, InfiniBand, Computer, Adoption, Copper, Marketing, Growth, Ecosystem, Industry, Association, QSFP-DD, TOP500, Compliance, ROCE, Trade association, Ethernet Alliance, Partnership, RDMA, NICS, Environment, University, LINPACK, NVIDIA, Name, OSFP, Data center bridging, NDR, HPE, QSFP, Intel, IBM, Leadership, IBTA, Public relations, Cassette tape, Risk management

InfiniBand also connects 62 of the Top100, making it the most used high-speed interconnect of the worlds top supercomputers.

Key Points: 
  • InfiniBand also connects 62 of the Top100, making it the most used high-speed interconnect of the worlds top supercomputers.
  • Record Turnout for 39th InfiniBand and RoCE Plugfest, Featuring NDR 400Gb/s InfiniBand and 400GbE
    This years IBTA Plugfest was the first testing instance of NDR 400Gb/s InfiniBand and 400GbE products, paving the way for the roll-out of the fastest interconnected systems in 2022.
  • Each product that completes compliance and interoperability testing is compiled into the resulting InfiniBand Integrators List and RoCE Integrators List.
  • The InfiniBand Trade Association was founded in 1999 and is chartered with maintaining and furthering the InfiniBand and the RoCE specifications.

Clever Cloud Selects French Kalray for Its High-performance Storage Solutions to Optimize Its New Data Center

Retrieved on: 
Tuesday, July 5, 2022
Semiconductor, Data Management, Apps, Applications, Technology, Mobile, Wireless, Software, Heart, Data-rate units, CEO, Euronext Growth, Renault–Nissan–Mitsubishi Alliance, Central processing unit, AXA, Massively parallel, Platform as a service, Massively parallel processor array, Santa Clara Convention Center, McDonald's, TBWA, Automation, Philosophy, RDMA, Partnership, Data, Central, SSD, PCI Express, Data processing unit, Software, Name, NVM Express, CPU, Data center bridging, CA, MAIF, Edge computing, CTO, Solution, NXP Semiconductors, Technology, CEA, AI, Acceleration, Solid-state drive, Macdonald, DPU, Multimedia, SNCF, Video game console, Cloud, Kalray, Airbus

A unique concept on the market today, the Flashbox as proposed by Kalray benefits from ultra-fast interfaces and low power consumption.

Key Points: 
  • A unique concept on the market today, the Flashbox as proposed by Kalray benefits from ultra-fast interfaces and low power consumption.
  • The Flashbox is currently being tested in Clever Cloud's infrastructure and will join its new Paris data center where it will work alongside the latest generation of compute servers Clever Cloud is installing there.
  • It will enable Clever Cloud to offer greater flexibility to its customers in managing storage, but also to support the launch of new offerings.
  • We are proud to be able to build innovative French storage solutions with Clever Cloud and thus contribute to the technological sovereignty of France and Europe, declares Eric Baissus, President and CEO of Kalray.

Huawei CloudFabric 3.0 Hyper-Converged DCN Solution Navigates the Green Computing Power Era with Higher-Quality Networks

Retrieved on: 
Thursday, April 28, 2022
Carbon, Transfer, Cloud, InfiniBand, Cloud computing, HAS, Speech, SuperPower, HPC, Efficiency, Organization, Data center bridging, Communication, Goal, DCN, Research, Supercooling, White paper, Industry, Mobile phone, Green, Huawei

As the digital economy enters the computing power era and global computing power initiatives such as China's "East-to-West Computing Resource Transfer" project get into full swing, the construction scale of data centers experiences explosive growth.

Key Points: 
  • As the digital economy enters the computing power era and global computing power initiatives such as China's "East-to-West Computing Resource Transfer" project get into full swing, the construction scale of data centers experiences explosive growth.
  • Leveraging lossless Ethernet networks, Huawei's CloudFabric 3.0 Hyper-Converged DCN Solution enhances network capabilities to maximize computing efficiency while minimizing the energy consumption per unit of computing power, thereby enabling green data center networks with optimal computing power.
  • In this way, it builds a large-scale, low-latency, and high-throughput computing power network, unleashing 100% of the computing power.
  • "The lossless network powered by our solution improves computing power while reducing energy consumption, fully aligning with the trends of data centers in the computing power era.

NVIDIA Announces Spectrum High-Performance Data Center Networking Infrastructure Platform

Retrieved on: 
Tuesday, March 22, 2022
Quanta, Omniverse, Data center bridging, NASDAQ, OVX, StackPath, OpenShift, IBM, Network, NVIDIA, Tenant, Automation, Growth, Transport, Palo Alto Networks, Ownership, Atos, RDMA, Baidu, GPU, Red Hat, WEKA, Launch pad, OVHcloud, DPU, Hewlett Packard Enterprise, Artificial intelligence, Dell Technologies, VMware, IEEE 802.1AE, Akamai Technologies, ASIC, Time, Architecture, Supermicro, GLOBE, Converge, Solution, GTC, Â, Bluefield, DDN, Efficiency, NetApp, AI, VAST Data, ASUS, Data, Cryptocurrency, Video game console, Lenovo, Canonical, Spectrum, Inspur

The worlds first 400Gbps end-to-end networking platform, NVIDIA Spectrum-4 provides 4x higher switching throughput than previous generations, with 51.2 terabits per second.

Key Points: 
  • The worlds first 400Gbps end-to-end networking platform, NVIDIA Spectrum-4 provides 4x higher switching throughput than previous generations, with 51.2 terabits per second.
  • It consists of the NVIDIA Spectrum-4 switch family , NVIDIA ConnectX-7 SmartNIC , NVIDIA BlueField-3 DPU and the DOCA data center infrastructure software to supercharge cloud-native applications at scale.
  • A new era of massive-scale cloud technologies, such as Omniverse, requires a transformation of data center architecture, said Kevin Deierling, vice president of Networking at NVIDIA.
  • The Spectrum platform accelerates state-of-the-art data center networking infrastructure for industry-leading customers and software vendors that include Akamai, Baidu, Canonical, Criteo, DDN, F5, Kuaishou, NetApp, Nutanix, OVHcloud, Pure Storage, Pluribus Networks, Red Hat, StackPath, VAST Data, VMware, WEKA and others.

Habana Labs Announces Turnkey AI Training Solution Featuring Habana Gaudi Platform and DDN AI400X2 Storage System

Retrieved on: 
Tuesday, November 16, 2021
Efficiency, NVM Express, RDMA, ROCE, View, Industry, PCI Express, Intel, TensorFlow, TB, Data management, Automation, GitHub, Machine learning, DDN, Multimedia, Solution, GB, HBM2, Data center bridging, IDC, Open Compute Project, Gigabit, NAS, Marketing, Medical device, Online shopping, Video game console

TEL-AVIV, Israel and SANTA CLARA, Calif., Nov. 16, 2021 /PRNewswire/ -- Habana Labs ,an Intel Company and leading developer of AI processors, today announced the availability of a turnkey, enterprise-class AI training solution featuring the Supermicro X12 Gaudi AI Training Server with the DDN AI400X2 Storage system.

Key Points: 
  • TEL-AVIV, Israel and SANTA CLARA, Calif., Nov. 16, 2021 /PRNewswire/ -- Habana Labs ,an Intel Company and leading developer of AI processors, today announced the availability of a turnkey, enterprise-class AI training solution featuring the Supermicro X12 Gaudi AI Training Server with the DDN AI400X2 Storage system.
  • With eight Habana Gaudi purpose-built AI processors, the Supermicro X12 Gaudi AI Server provides customers with highly cost-efficient AI training, ease of use and system scalability.
  • The turnkey AI training solution comespre-configured withone, two and four X12 server options to address AI training capacity requirements.
  • To learn more about the high-efficiency combination of the Supermicro X12 Gaudi AI Server and DDN AI400X2 appliance, see the DDN whitepaper: DDN A3I Solutions with Supermicro X12 Gaudi AI Servers .

Oracle Introduces Next-Generation Exadata X9M Platforms

Retrieved on: 
Tuesday, September 28, 2021
Zero, Cloud, Microsoft Azure SQL Database, Oracle Zero Data Loss Recovery Appliance, ACL, Global, Partnership, Azure, Data center, IOPS, Access control, TB, Data center bridging, Running, Efficiency, Memory, Amazon Relational Database Service, ROCE, AWS, Oracle Streams, Microsoft, 3D XPoint, Amazon, SQL, RDS, RDMA, Third, Siebel Systems, IDC, SEC, Central processing unit, OLTP, Siebel, Algorithm, Access-control list, Oracle Database Appliance, DBT Online Inc., Equinix, Intel, IO, Machine learning, Oracle Database, Goldengate, Database, Computer data storage

AUSTIN, Texas, Sept. 28, 2021 /PRNewswire/ -- Oracle today announced the availability of the Oracle Exadata X9M platforms , the latest version of the industry's fastest and most affordable systems for running Oracle Database.

Key Points: 
  • AUSTIN, Texas, Sept. 28, 2021 /PRNewswire/ -- Oracle today announced the availability of the Oracle Exadata X9M platforms , the latest version of the industry's fastest and most affordable systems for running Oracle Database.
  • The new Exadata X9M offerings include Oracle Exadata Database Machine X9M and Exadata Cloud@Customer X9Mthe only platform that runs Oracle Autonomous Database in customer data centers.
  • Next-generation Exadata X9M platforms accelerate online transaction processing (OLTP) with more than 70 percent higher IOPs rates and IO latencies of under 19 microseconds.
  • "The Oracle Exadata X9M generation continues our strategy of providing customers exceptional value by delivering the world's fastest and most available Oracle Database platform, and making it available everywherein the public cloud, cloud@customer, and on-premises," said Juan Loaiza, executive vice president, Mission-Critical Database Technologies, Oracle.

Huawei and Third-Party Organizations Reach a Consensus: Lossless Ethernet-based NVMe over Fabric (NoF) Is the Inevitable Choice in the All-Flash Era

Retrieved on: 
Friday, September 24, 2021
Dorado, Huawei, Interop, TCP, IOPS, Data center bridging, ROCE, Engineering, CEO, NVM Express, Bank of China, SCSI, Packet loss, Growth, IDC, FC, Founder, Remote direct memory access, Acceleration, Bank, International Data Group, Founder CEO, Industry, Mobile phone, Computer data storage

It was at the session that all parties reached a conclusion that NVMe over Fabric is the inevitable choice in the all-flash era.

Key Points: 
  • It was at the session that all parties reached a conclusion that NVMe over Fabric is the inevitable choice in the all-flash era.
  • More importantly, Huawei proposed the Lossless Ethernet Storage Network Solution NoF+, otherwise known as NoF+ Solution, an industry-changing innovation that is designed to unlock the potential of all-flash storage.
  • In addition, banks everywhere now need to provide 24/7 services, which presents unprecedented challenges to transaction latency and reliability.
  • It takes full advantage of the million-level IOPS of all-flash storage and meets the petabyte-level development requirements of the financial industry.