Intel this week announced details of its new Blockscale ASIC chip designed specifically for more efficient blockchain computing than CPUs or GPUs. It first said it was making such a chip just two months ago.Blockscale is specifically designed to process the Secure Hash Algorithm-256 (SHA-256), which is used by blockchain, and the performance is phenomenal, at least on paper. Blockscale has a hash rate operating speed of up to 580GH/s, or gigahashes per second.To read this article in full, please click here
The conventional wisdom is that you should update your IT gear, namely the servers, every three-to-five years, which is usually when service warranties run out. However, some companies hold onto their gear for longer than that for a variety of reasons: lack of funds, business uncertainty, on-premises versus the cloud, and so forth.And for a while, the CPU guys were helping. New generations of processors were only incrementally faster than the old ones making it hard to justify an upgrade. The result was longer lifecycles for server hardware. A 2020 survey by IDC found 20.3% of respondents holding on to servers for six years and 12.4% keeping servers for seven years or more.To read this article in full, please click here
The conventional wisdom is that you should update your IT gear, namely the servers, every three-to-five years, which is usually when service warranties run out. However, some companies hold onto their gear for longer than that for a variety of reasons: lack of funds, business uncertainty, on-premises versus the cloud, and so forth.And for a while, the CPU guys were helping. New generations of processors were only incrementally faster than the old ones making it hard to justify an upgrade. The result was longer lifecycles for server hardware. A 2020 survey by IDC found 20.3% of respondents holding on to servers for six years and 12.4% keeping servers for seven years or more.To read this article in full, please click here
The old saying “adversity makes for strange bedfellows” has been proven true, with Nvidia saying it is now willing to work with Intel’s foundry business to manufacture its chips.Nvidia CEO Jen-Hsun Huang dropped the news on a press call when he was asked . about diversifying the company’s supply chain, which relies on TSMC for its chip manufacturing, and TSMC is both overloaded with orders and in a politically unstable region of the world (Taiwan).Huang said his company realized it needed more resilience going forward, and so over the last couple years has added to the number of process nodes it uses, and is in more fabs than ever. “So we've expanded our supply chain, supply base, probably four-fold in the last two years,” Huang said.To read this article in full, please click here
The old saying “adversity makes for strange bedfellows” has been proven true, with Nvidia saying it is now willing to work with Intel’s foundry business to manufacture its chips.Nvidia CEO Jen-Hsun Huang dropped the news on a press call when he was asked . about diversifying the company’s supply chain, which relies on TSMC for its chip manufacturing, and TSMC is both overloaded with orders and in a politically unstable region of the world (Taiwan).Huang said his company realized it needed more resilience going forward, and so over the last couple years has added to the number of process nodes it uses, and is in more fabs than ever. “So we've expanded our supply chain, supply base, probably four-fold in the last two years,” Huang said.To read this article in full, please click here
At its GPU technology conference (GTC) last year, Nvidia announced it would come out with its own server chip called Grace based on the Arm Neoverse v9 server architecture. At the time, details were scant, but this week Nvidia revealed the details, and they are remarkable.With Grace, customers have two options, both dubbed superchips by Nvidia. The first is the Grace Hopper Superchip that was formally introduced last year, but only broadly described. It consists of a 72-core CPU, and a Hopper H100 GPU tightly connected by Nvidia’s new high-speed NVLink-C2C chip-to-chip interconnect, which has 900GB/s of transfer speed.To read this article in full, please click here
At its GPU technology conference (GTC) last year, Nvidia announced it would come out with its own server chip called Grace based on the Arm Neoverse v9 server architecture. At the time, details were scant, but this week Nvidia revealed the details, and they are remarkable.With Grace, customers have two options, both dubbed superchips by Nvidia. The first is the Grace Hopper Superchip that was formally introduced last year, but only broadly described. It consists of a 72-core CPU, and a Hopper H100 GPU tightly connected by Nvidia’s new high-speed NVLink-C2C chip-to-chip interconnect, which has 900GB/s of transfer speed.To read this article in full, please click here
Nvidia is known for its GPUs, but has introduced Spectrum-4, a combination of networking technologies that reinforces its commitment not only to graphics processors, but also to systems designed to handle the demanding network workloads of AI and high-performance computing.The latest Nvidia Spectrum products rely on the new Spectrum-4 Ethernet-switch ASIC that boasts 51.2 Tb/s switching and routing capacity. The chip underpins the latest members of the company’s Spectrum switches, which are available later this year. The switches are part of a larger Spectrum-4 platform that integrates Nvidia’s ConnectX-7 smartNIC, its new BlueField-3 DPU, and its DOCA software-development platform.To read this article in full, please click here
Nvidia is known for its GPUs, but has introduced Spectrum-4, a combination of networking technologies that reinforces its commitment not only to graphics processors, but also to systems designed to handle the demanding network workloads of AI and high-performance computing.The latest Nvidia Spectrum products rely on the new Spectrum-4 Ethernet-switch ASIC that boasts 51.2 Tb/s switching and routing capacity. The chip underpins the latest members of the company’s Spectrum switches, which are available later this year. The switches are part of a larger Spectrum-4 platform that integrates Nvidia’s ConnectX-7 smartNIC, its new BlueField-3 DPU, and its DOCA software-development platform.To read this article in full, please click here
While the rest of the computing industry struggles to get to one exaflop of computing, Nvidia is about to blow past everyone with an 18-exaflop supercomputer powered by a new GPU architecture.The H100 GPU, has 80 billion transistors (the previous generation, Ampere, had 54 billion) with nearly 5TB/s of external connectivity and support for PCIe Gen5, as well as High Bandwidth Memory 3 (HBM3), enabling 3TB/s of memory bandwidth, the company says. Due out in the third quarter, it’s the first in a new family of GPUs named Hopper after Admiral Grace Hopper who created COBOL and coined the term computer bug.To read this article in full, please click here
While the rest of the computing industry struggles to get to one exaflop of computing, Nvidia is about to blow past everyone with an 18-exaflop supercomputer powered by a new GPU architecture.The H100 GPU, has 80 billion transistors (the previous generation, Ampere, had 54 billion) with nearly 5TB/s of external connectivity and support for PCIe Gen5, as well as High Bandwidth Memory 3 (HBM3), enabling 3TB/s of memory bandwidth, the company says. Due out in the third quarter, it’s the first in a new family of GPUs named Hopper after Admiral Grace Hopper who created COBOL and coined the term computer bug.To read this article in full, please click here
This year, server vendors will begin shifting to a new form of memory, Double Data Rate version 5, or DDR5 for short. With its improved performance, it will be very appealing in certain use cases, like virtualization and artificial intelligence. We’ll get to that in a minute.The DDR spec has been developed by the Joint Electronic Device Engineering Council since 2001, and with each iteration the spec supports faster speed and lower power draw. This holds true for DDR5.
[ Get regularly scheduled insights by signing up for Network World newsletters. ]To read this article in full, please click here
This year, server vendors will begin shifting to a new form of memory, Double Data Rate version 5, or DDR5 for short. With its improved performance, it will be very appealing in certain use cases, like virtualization and artificial intelligence. We’ll get to that in a minute.The DDR spec has been developed by the Joint Electronic Device Engineering Council since 2001, and with each iteration the spec supports faster speed and lower power draw. This holds true for DDR5.
[ Get regularly scheduled insights by signing up for Network World newsletters. ]To read this article in full, please click here
In the battle between Intel and AMD, it can be easy to overlook Marvell Technology, but you shouldn’t. Through acquisition and organic growth, the company has turned into quite a powerhouse playing in multiple areas.Marvell is the first major vendor to support the Arm on 5G initiative that Arm unveiled last October, when it launched the Arm 5G Solutions Lab. The lab is designed to let hardware and software partners work on 5G-based products running on Arm architecture.
Read more: SmartNICs set to infiltrate enterprise networksTo read this article in full, please click here
Supply chain problems for Wi-Fi 6E access points are so bad that enterprises are skipping that version of wireless technology and waiting until Wi-Fi 7 equipment starts to ship late next year, says market researcher Dell'Oro Group.Wi-Fi 6E builds on Wi-Fi 6 by adding the the 6GHz band (5.925 GHz to 7.125 GHz), where, currently, there is a lot less traffic and much lower latency than in 2.4GHz and 5GHz bands that Wi-Fi 6. That extra bandwidth makes 6E a logical choice for latency-sensitive applications.But you can’t use something if you can’t buy it, and Dell'Oro says that based on its discussions with enterprises, 6E products are in very limited supply or unavailable.To read this article in full, please click here
Supply chain problems for Wi-Fi 6E access points are so bad that enterprises are skipping that version of wireless technology and waiting until Wi-Fi 7 equipment starts to ship late next year, says market researcher Dell'Oro Group.Wi-Fi 6E builds on Wi-Fi 6 by adding the the 6GHz band (5.925 GHz to 7.125 GHz), where, currently, there is a lot less traffic and much lower latency than in 2.4GHz and 5GHz bands that Wi-Fi 6. That extra bandwidth makes 6E a logical choice for latency-sensitive applications.But you can’t use something if you can’t buy it, and Dell'Oro says that based on its discussions with enterprises, 6E products are in very limited supply or unavailable.To read this article in full, please click here
Verizon Business has added VMware to its global managed-SD-WAN portfolio as part of its Network as a Service (NaaS) strategy. Verizon made the announcement at the Mobile World Conference event in Barcelona.Verizon’s Managed SD-WAN is designed for hybrid-cloud environments and uses application-aware routing to make sure customer data takes the right path to its destination. This allows customers to use their private network for demanding, latency-sensitive apps while sending less critical data over public networks.VMware SD-WAN features orchestration around centralized policy, monitoring, reporting, and analytics via Verizon Enterprise Center. It also offers SD WAN gateways with controllers. VMware Gateways are points of presencelocated around the world to provide physically close, low-latency connectivity to customer edge devices.To read this article in full, please click here
Dell's newest entry-level block storage array is the PowerVault ME5 series, aimed at price-sensitive customers with a focus on ease of deployment and affordability.The array's predecessor, the PowerVault ME4, was released in 2018. So it was overdue for an upgrade – and Dell delivered.The ME5 features significant performance and capacity improvements compared to the ME4. Between the hardware and software upgrades, Dell says the PowerVault ME5 offers twice the performance, throughput, capacity and memory of the ME4. The ME5 has newer Xeon processors with twice as many cores as the ME4, and controller memory has been increased to 16GB per controller.To read this article in full, please click here
Dell’s newest entry-level block-storage array is the PowerVault ME5 series, aimed at price-sensitive customers with a focus on ease of deployment and affordability.The array’s predecessor, the PowerVault ME4, was released in 2018. So it was overdue for an upgrade—and Dell delivered.The ME5 features significant performance and capacity improvements compared to the ME4. Between the hardware and software upgrades, Dell says the PowerVault ME5 offers twice the performance, throughput, capacity and memory of the ME4. The ME5 has newer Xeon processors with twice as many cores as the ME4, and controller memory has been increased to 16GB per controller.To read this article in full, please click here
How bad is the chip supply shortage? Gartner reports that clients are complaining about lead times as long as 400 days to get networking equipment, plus pricing increases and missed ship dates.“We expect lead times to remain high through early 2023, at which point we expect slow incremental improvement over the course of months,” Gartner wrote in a report titled, "What Are My Options for Dealing With Long Lead Times on Network Equipment?"
Read more: Chip shortage has networking vendors scramblingTo read this article in full, please click here
Andy Patrizio