Andy Patrizio

Author Archives: Andy Patrizio

Liquid cooling and edge computing are featured at Gartner show

Research firm Gartner is holding its IT Infrastructure, Operations, and Cloud Strategies Conference (IOCS) in Las Vegas this week, and a few news announcements from the show give an indication as to where data-center technology is headed.First up, Schneider Electric and Iceotope formally introduced their integrated rack with chassis-based immersive liquid-cooling designs. The deal was announced in October but now the details are out. In addition to Schneider and Iceotope, the alliance also includes Avnet, an electronic-component distributor.To read this article in full, please click here

Liquid cooling and edge computing are featured at Gartner show

Research firm Gartner is holding its IT Infrastructure, Operations, and Cloud Strategies Conference (IOCS) in Las Vegas this week, and a few news announcements from the show give an indication as to where data-center technology is headed.First up, Schneider Electric and Iceotope formally introduced their integrated rack with chassis-based immersive liquid-cooling designs. The deal was announced in October but now the details are out. In addition to Schneider and Iceotope, the alliance also includes Avnet, an electronic-component distributor.To read this article in full, please click here

Google Cloud bare-metal initiative targets migrating legacy apps from on-prem

In the cloud-services market, bare metal offerings have lagged behind virtualized ones, mostly because the use of the cloud for things like elastic apps and developer environments are better suited to instances with a native operating system.The term “bare metal” simply means no software of any kind, not even a hypervisor. Customers provide their own operating environments, and the provider offers nothing more than CPUs, memory, and storage. Up to now, IBM has led the charge with bare-metal services because SoftLayer, the major data-center provider it acquired in 2014, was heavily involved in that business.To read this article in full, please click here

Data centers in 2020: Automation, cheaper memory

It’s that time of year again when those of us in the press make our annual prognostications for the coming year. Some things we saw coming; the rise of the cloud and the advance of SSD. Others, like the return of many cloud migrations to on-premises or the roaring comeback of AMD, went right by us. We do our best but occasionally there are surprises.So with that, let’s take a peek into the always cloudy (no pun intended) crystal ball and make 10 data-center-oriented predictions.IoT spawns data-center growth in urban areas This isn’t a hard prediction to make since it’s already happening. For the longest time, data centers were placed in the middle of nowhere near renewable energy (usually hydro), but need is going to force more expansion in major metro areas. IoT will be one driver but so will the increasing use of data center providers like Equinix and DRT as interconnection providers.To read this article in full, please click here

Data centers in 2020: Automation, cheaper memory

It’s that time of year again when those of us in the press make our annual prognostications for the coming year. Some things we saw coming; the rise of the cloud and the advance of SSD. Others, like the return of many cloud migrations to on-premises or the roaring comeback of AMD, went right by us. We do our best but occasionally there are surprises.So with that, let’s take a peek into the always cloudy (no pun intended) crystal ball and make 10 data-center-oriented predictions.IoT spawns data-center growth in urban areas This isn’t a hard prediction to make since it’s already happening. For the longest time, data centers were placed in the middle of nowhere near renewable energy (usually hydro), but need is going to force more expansion in major metro areas. IoT will be one driver but so will the increasing use of data center providers like Equinix and DRT as interconnection providers.To read this article in full, please click here

Seagate doubles HDD performance with multi-actuator technology

Seagate has taken the wraps off its Exos 2X14 enterprise hard drive. It's the first to integrate Seagate's MACH.2 multi-actuator technology, which is a method of turning one hard disk into two and doubling performance.The technology is pretty straightforward. Say you have four platters in a disk drive. The actuator controls the drive heads and moves them all in unison over all four platters. Seagate's multi-actuator makes two independent actuators out of one, so in a six-platter drive, the two actuators cover three platters each. READ MORE: SSD vs. HDD: Choosing between solid-state and hard-disk drivesTo read this article in full, please click here

Ampere preps an 80-core Arm processor for the cloud

Ampere Computing, the semiconductor startup led by former Intel president Renee James that designs Arm-based server processors, is preparing to launch its next-generation CPU by mid-2020.The upcoming chip will have 80 cores, much more than the 32-core processor the company shipped last year and vastly more than x86 CPUs by Intel and AMD. Ampere’s design is different. Instead of multiple threads per core, each core is single threaded.[Get regularly scheduled insights by signing up for Network World newsletters.] Jeff Wittich, Ampere’s senior vice president of products, said that was by design, to avoid some of the CPU vulnerabilities that crept into x86 chips but also to avoid the “noisy neighbor” problem in cloud service-provider networks.To read this article in full, please click here

Ampere preps an 80-core Arm processor for the cloud

Ampere Computing, the semiconductor startup led by former Intel president Renee James that designs Arm-based server processors, is preparing to launch its next-generation CPU by mid-2020.The upcoming chip will have 80 cores, much more than the 32-core processor the company shipped last year and vastly more than x86 CPUs by Intel and AMD. Ampere’s design is different. Instead of multiple threads per core, each core is single threaded.Jeff Wittich, Ampere’s senior vice president of products, said that was by design, to avoid some of the CPU vulnerabilities that crept into x86 chips but also to avoid the “noisy neighbor” problem in cloud service-provider networks.To read this article in full, please click here

HPE, DoE partner for AI-driven energy efficiency

HP Enterprise has partnered with the National Renewable Energy Laboratory (NREL), a unit of the Department of Energy, to create AI and machine learning-systems for greater data-center energy efficiency.The Department of Energy lab will provide HPE with multiple years’ worth of historical data from sensors within its supercomputers and in its Energy Systems Integration Facility (ESIF) High-Performance Computing (HPC) Data Center, one of the world's most efficient data centers. This information will help other organizations to optimize their own operations, said NREL.To read this article in full, please click here

HPE, DoE partner for AI-driven energy efficiency

HP Enterprise has partnered with the National Renewable Energy Laboratory (NREL), a unit of the Department of Energy, to create AI and machine learning-systems for greater data-center energy efficiency.The Department of Energy lab will provide HPE with multiple years’ worth of historical data from sensors within its supercomputers and in its Energy Systems Integration Facility (ESIF) High-Performance Computing (HPC) Data Center, one of the world's most efficient data centers. This information will help other organizations to optimize their own operations, said NREL.To read this article in full, please click here

Inphi launches DSPs for long-distance 400Gb networking

Inphi, a developer of high-speed data interconnects, announced it has begun sampling its new Canopus coherent digital signal processor (DSP), which it claims reduces power draw by up to 75 percent and triples the throughput of data over fiber networks, especially over long distances.The Canopus processor comes on a plug-in module about the size of a cigarette lighter that goes in existing networking equipment. The chip is built on a 7nm manufacturing process, and its silicon geometry delivers over 75 percent reduction in DSP power dissipation and size as compared to the current generation of coherent DSPs.Coherent optical transmission is a technique for transporting considerably more information through a fiber optic cable, and is especially popular when transporting over long distances. It uses modulation and phases of the light to amplify transmission. A DSP is often needed to manage and clean up the photonics.To read this article in full, please click here

Nvidia quietly unveils faster, lower power Tesla GPU accelerator

Nvidia was all over Supercomputing 19 last week, not surprisingly, and made a lot of news which we will get into later. But overlooked was perhaps the most interesting news of all: a new generation graphics-acceleration card that is faster and way more power efficient.Multiple attendees and news sites spotted it at the show, and Nvidia confirmed to me that this is indeed a new card. Nvidia’s “Volta” generation of Tesla GPU-accelerator cards has been out since 2017, so an upgrade was well overdue.[Get regularly scheduled insights by signing up for Network World newsletters.] The V100S comes only in PCI Express 3 form factor for now but is expected to eventually support Nvidia’s SXM2 interface. SXM is a dual-slot card design by Nvidia that requires no connection to the power supply, unlike the PCIe cards. SXM2 allows the GPU to communicate either with each other or to the CPU through Nvidia’s NVLink, a high-bandwidth, energy-efficient interconnect that can transfer data up to ten times faster than PCIe.To read this article in full, please click here

Nvidia quietly unveils faster, lower power Tesla GPU accelerator

Nvidia was all over Supercomputing 19 last week, not surprisingly, and made a lot of news which we will get into later. But overlooked was perhaps the most interesting news of all: a new generation graphics-acceleration card that is faster and way more power efficient.Multiple attendees and news sites spotted it at the show, and Nvidia confirmed to me that this is indeed a new card. Nvidia’s “Volta” generation of Tesla GPU-accelerator cards has been out since 2017, so an upgrade was well overdue.[Get regularly scheduled insights by signing up for Network World newsletters.] The V100S comes only in PCI Express 3 form factor for now but is expected to eventually support Nvidia’s SXM2 interface. SXM is a dual-slot card design by Nvidia that requires no connection to the power supply, unlike the PCIe cards. SXM2 allows the GPU to communicate either with each other or to the CPU through Nvidia’s NVLink, a high-bandwidth, energy-efficient interconnect that can transfer data up to ten times faster than PCIe.To read this article in full, please click here

8 ways to prepare your data center for AI’s power draw

As artificial intelligence takes off in enterprise settings, so will data center power usage. AI is many things, but power efficient is not one of them.For data centers running typical enterprise applications, the average power consumption for a rack is around 7 kW. Yet it’s common for AI applications to use more than 30 kW per rack, according to data center organization AFCOM. That’s because AI requires much higher processor utilization, and the processors – especially GPUs – are power hungry. Nvidia GPUs, for example, may run several orders of magnitude faster than a CPU, but they also consume twice as much power per chip. Complicating the issue is that many data centers are already power constrained.To read this article in full, please click here

8 ways to prepare your data center for AI’s power draw

As artificial intelligence takes off in enterprise settings, so will data center power usage. AI is many things, but power efficient is not one of them.For data centers running typical enterprise applications, the average power consumption for a rack is around 7 kW. Yet it’s common for AI applications to use more than 30 kW per rack, according to data center organization AFCOM. That’s because AI requires much higher processor utilization, and the processors – especially GPUs – are power hungry. Nvidia GPUs, for example, may run several orders of magnitude faster than a CPU, but they also consume twice as much power per chip. Complicating the issue is that many data centers are already power constrained.To read this article in full, please click here

Cumulus Networks updates its network-centric Linux distribution

The Linux distribution ecosystem is pretty set, with Red Hat and Canonical in the leadership positions, followed closely by SuSe and home brews from the likes of IBM and other major vendors. Even Microsoft has its own distro for Azure users.And then there is Cumulus Networks, which specializes in networking software. It just released Cumulus Linux 4.0 and NetQ 2.4, its cloud network deployment and management console. With this release, Cumulus is claiming its Linux is its most stable and reliable software stack yet and NetQ is the most comprehensive end-to-end network automation product.To read this article in full, please click here

Intel targets Nvidia (again) with GPU and cross-processor API

Third time’s the charm? Intel is hoping so. It released details of its Xe Graphics Architecture, with which it plans to span use cases from mobility to high-performance computing (HPC) servers – and which it hopes will succeed where its Larrabee GPU and Xeon Phi manycore processors failed.It’s no secret Intel wants a piece of the high-performance computing HPC action, given that it introduced the chip and other products at it Intel HPC Developer Conference in Denver, Colo., this week just ahead of the Supercomputing ’19 tradeshow.To read this article in full, please click here

Intel targets Nvidia (again) with GPU and cross-processor API

Third time’s the charm? Intel is hoping so. It released details of its Xe Graphics Architecture, with which it plans to span use cases from mobility to high-performance computing (HPC) servers – and which it hopes will succeed where its Larrabee GPU and Xeon Phi manycore processors failed.It’s no secret Intel wants a piece of the high-performance computing HPC action, given that it introduced the chip and other products at it Intel HPC Developer Conference in Denver, Colo., this week just ahead of the Supercomputing ’19 tradeshow.To read this article in full, please click here

Digital Realty jumps into interconnection business

Data center provider Digital Realty Trust isn't resting after its massive EMEA push via the acquisition of Interxion. The company unveiled PlatformDIGITAL, an initiative designed to provide interconnections to customers and manage big data. Digital Realty made the announcement at its MarketplaceLIVE conference. At the heart of the PlatformDIGITAL strategy is Pervasive Datacenter Architecture (PDx), which offers "fit-for-purpose" data center designs meant to solve scale, configuation and connectivity issues faced by enterprise colocation customers.To read this article in full, please click here

Digital Realty jumps into interconnection business

Data center provider Digital Realty Trust isn't resting after its massive EMEA push via the acquisition of Interxion. The company unveiled PlatformDIGITAL, an initiative designed to provide interconnections to customers and manage big data. Digital Realty made the announcement at its MarketplaceLIVE conference. At the heart of the PlatformDIGITAL strategy is Pervasive Datacenter Architecture (PDx), which offers "fit-for-purpose" data center designs meant to solve scale, configuation and connectivity issues faced by enterprise colocation customers.To read this article in full, please click here

1 2 3 48