Andy Patrizio
Author Archives: Andy Patrizio
- Home → Author's archive:
Andy Patrizio
0
AMD hosts an Epyc party — and everyone wants in
Last week, AMD launched the second generation of its Epyc server processor, the Epyc 7002 series a.k.a. “Rome,” and it’s a far cry from the days when it held a release party for the Opteron and no one showed up. These days, AMD has a whole lot of friends.Of course, it helps to deliver a part people want, and it looks like the Epyc 7002 is all that. It builds considerably upon the first generation, code-named “Naples,” delivered two years ago. One chip packs up to 64 cores and two threads per core, double the max of 32 cores in Naples. It has eight memory channels and up to 128 lanes of PCI Express Gen 4.[ Also read: What is quantum computing (and why enterprises should care) ] The Epyc 7002 achieves this massive core count through “chiplets,” eight small chips in the CPU die with eight cores each and connected by a high-speed interconnect. A single monolithic 64-core die is impractical from a manufacturing standpoint. There is so much more that can go wrong with 64 cores than 16. Plus, AMD is manufacturing this on a 7nm process (Intel is just getting to 10nm), so Continue reading
0
AMD hosts an Epyc party — and everyone wants in
Last week, AMD launched the second generation of its Epyc server processor, the Epyc 7002 series a.k.a. “Rome,” and it’s a far cry from the days when it held a release party for the Opteron and no one showed up. These days, AMD has a whole lot of friends.Of course, it helps to deliver a part people want, and it looks like the Epyc 7002 is all that. It builds considerably upon the first generation, code-named “Naples,” delivered two years ago. One chip packs up to 64 cores and two threads per core, double the max of 32 cores in Naples. It has eight memory channels and up to 128 lanes of PCI Express Gen 4.[ Also read: What is quantum computing (and why enterprises should care) ] The Epyc 7002 achieves this massive core count through “chiplets,” eight small chips in the CPU die with eight cores each and connected by a high-speed interconnect. A single monolithic 64-core die is impractical from a manufacturing standpoint. There is so much more that can go wrong with 64 cores than 16. Plus, AMD is manufacturing this on a 7nm process (Intel is just getting to 10nm), so Continue reading
0
Xilinx launches new FPGA cards that can match GPU performance
Xilinx has launched a new FPGA card, the Alveo U50, that it claims can match the performance of a GPU in areas of artificial intelligence (AI) and machine learning.The company claims the card is the industry’s first low-profile adaptable accelerator with PCIe Gen 4 support, which offers double the throughput over PCIe Gen3. It was finalized in 2017, but cards and motherboards to support it have been slow to come to market.The Alveo U50 provides customers with a programmable low-profile and low-power accelerator platform built for scale-out architectures and domain-specific acceleration of any server deployment, on premises, in the cloud, and at the edge.To read this article in full, please click here
0
Xilinx launches new FPGA cards that can match GPU performance
Xilinx has launched a new FPGA card, the Alveo U50, that it claims can match the performance of a GPU in areas of artificial intelligence (AI) and machine learning.The company claims the card is the industry’s first low-profile adaptable accelerator with PCIe Gen 4 support, which offers double the throughput over PCIe Gen3. It was finalized in 2017, but cards and motherboards to support it have been slow to come to market.The Alveo U50 provides customers with a programmable low-profile and low-power accelerator platform built for scale-out architectures and domain-specific acceleration of any server deployment, on premises, in the cloud, and at the edge.To read this article in full, please click here
0
Intel and Lenovo partner on HPC and AI initiatives
Intel and Lenovo this week announced a new partnership aimed at accelerating artificial intelligence (AI) and high-performance computing (HPC) products by bringing together their respective technologies.The collaboration will integrate Lenovo's TruScale Infrastructure and Lenovo Neptune liquid cooling technology with a variety of Intel technologies, including its Optane DC persistent memory, Intel oneAPI programming framework, and current and future generations of its Xeon Scalable processors. Read more data center stories NVMe over Fabrics creates data-center storage disruption How AI can improve network capacity planning HPE to buy Cray, offer HPC as a service Data center workloads become more complex How to get a handle on multicloud management TruScale is a consumption-based offering that allows customers to use on-premises data center hardware and services without having to purchase the equipment outright; enterprises pay for the use, and Lenovo monitors their activity. It's a model all of the major OEMs have adopted in response to the cloud.To read this article in full, please click here
0
Intel and Lenovo partner on HPC and AI initiatives
Intel and Lenovo this week announced a new partnership aimed at accelerating artificial intelligence (AI) and high-performance computing (HPC) products by bringing together their respective technologies.The collaboration will integrate Lenovo's TruScale Infrastructure and Lenovo Neptune liquid cooling technology with a variety of Intel technologies, including its Optane DC persistent memory, Intel oneAPI programming framework, and current and future generations of its Xeon Scalable processors. Read more data center stories NVMe over Fabrics creates data-center storage disruption How AI can improve network capacity planning HPE to buy Cray, offer HPC as a service Data center workloads become more complex How to get a handle on multicloud management TruScale is a consumption-based offering that allows customers to use on-premises data center hardware and services without having to purchase the equipment outright; enterprises pay for the use, and Lenovo monitors their activity. It's a model all of the major OEMs have adopted in response to the cloud.To read this article in full, please click here
0
Intel pulls the plug on Omni-Path networking fabric architecture
Intel’s battle to gain ground in the high-performance computing (HPC) market isn’t going so well. The Omni-Path Architecture it had pinned its hopes on has been scrapped after one generation.An Intel spokesman confirmed to me that the company will no longer offer Intel OmniPath Architecture 200 (OPA200) products to customers, but it will continue to encourage customers, OEMs, and partners to use OPA100 in new designs. “We are continuing to sell, maintain, and support OPA100. We actually announced some new features for OPA100 back at International Supercomputing in June,” the spokesperson said via email.[ Learn who's developing quantum computers. ] Intel said it continues to invest in connectivity solutions for its customers and that the recent acquisition of Barefoot Networks is an example of Intel’s strategy of supporting end-to-end cloud networking and infrastructure. It would not say if Barefoot’s technology would be the replacement for OPA.To read this article in full, please click here
0
Intel pulls the plug on Omni-Path networking fabric architecture
Intel’s battle to gain ground in the high-performance computing (HPC) market isn’t going so well. The Omni-Path Architecture it had pinned its hopes on has been scrapped after one generation.An Intel spokesman confirmed to me that the company will no longer offer Intel OmniPath Architecture 200 (OPA200) products to customers, but it will continue to encourage customers, OEMs, and partners to use OPA100 in new designs. “We are continuing to sell, maintain, and support OPA100. We actually announced some new features for OPA100 back at International Supercomputing in June,” the spokesperson said via email.[ Learn who's developing quantum computers. ] Intel said it continues to invest in connectivity solutions for its customers and that the recent acquisition of Barefoot Networks is an example of Intel’s strategy of supporting end-to-end cloud networking and infrastructure. It would not say if Barefoot’s technology would be the replacement for OPA.To read this article in full, please click here
0
Is your enterprise software committing security malpractice?
Back when this blog was dedicated to all things Microsoft I routinely railed against the spying aspects of Windows 10. Well, apparently that’s nothing compared to what enterprise security, analytics, and hardware management tools are doing.An analytics firm called ExtraHop examined the networks of its customers and found that their security and analytic software was quietly uploading information to servers outside of the customer's network. The company issued a report and warning last week.ExtraHop deliberately chose not to name names in its four examples of enterprise security tools that were sending out data without warning the customer or user. A spokesperson for the company told me via email, “ExtraHop wants the focus of the report to be the trend, which we have observed on multiple occasions and find alarming. Focusing on a specific group would detract from the broader point that this important issue requires more attention from enterprises.”To read this article in full, please click here
0
Is your enterprise software committing security malpractice?
Back when this blog was dedicated to all things Microsoft I routinely railed against the spying aspects of Windows 10. Well, apparently that’s nothing compared to what enterprise security, analytics, and hardware management tools are doing.An analytics firm called ExtraHop examined the networks of its customers and found that their security and analytic software was quietly uploading information to servers outside of the customer's network. The company issued a report and warning last week.ExtraHop deliberately chose not to name names in its four examples of enterprise security tools that were sending out data without warning the customer or user. A spokesperson for the company told me via email, “ExtraHop wants the focus of the report to be the trend, which we have observed on multiple occasions and find alarming. Focusing on a specific group would detract from the broader point that this important issue requires more attention from enterprises.”To read this article in full, please click here
0
Xilinx closes SolarFlare purchase, promises high-performance networking
Network acceleration appears to be all the rage these days, what with Nvidia acquiring Mellanox, the advent of High Bandwidth Memory 2E targeting networking chips, and now Xilinx closing of its acquisition of low-latency network provider SolarFlare.SolarFlare makes a high-speed network interface card (NIC) using field-programmable gate arrays (FPGAs) to build SmartNICs sold under the X2 brand. These PCI Express network interface cards run network, storage, and compute acceleration, offloading that work from the CPU. SolarFlare also develops application acceleration software to fully utilize the cards.To read this article in full, please click here
0
VMware’s Bitfusion acquisition could be a game-changer for GPU computing
In a low-key move that went under the radar of a lot of us, last week VMware snapped up a startup called Bitfusion, which makes virtualization software for accelerated computing. It improves performance of virtual machines by offloading processing to accelerator chips, such as GPUs, FPGAs, or other custom ASICs.Bitfusion provides sharing of GPU resources among isolated GPU compute workloads, allowing workloads to be shared across the customer’s network. This way workloads are not tied to one physical server but shared as a pool of resources, and if multiple GPUs are brought to bear, performance naturally increases.“In many ways, Bitfusion offers for hardware acceleration what VMware offered to the compute landscape several years ago. Bitfusion also aligns well with VMware’s ‘Any Cloud, Any App, Any Device’ vision with its ability to work across AI frameworks, clouds, networks, and formats such as virtual machines and containers,” said Krish Prasad, senior vice president and general manager of the Cloud Platform Business Unit at VMware, in a blog post announcing the deal.To read this article in full, please click here
0
NVMe over Fabrics enterprise storage spec enters final review process
NVM Express Inc., the developer of the NVMe spec for enterprise SSDs, announced that its NVMe-oF architecture has entered a final 45-day review, an important step toward release of a formal specification for enterprise SSD makers.NVMe-oF stands for NVMe over Fabrics, a mechanism to transfer data between a host computer and a target SSD or system over a network, such as Ethernet, Fibre Channel (FC), or InfiniBand. NVM Express first released the 1.0 spec of NVMe-oF in 2016, so this is long overdue.To read this article in full, please click here
0
Verizon launches interconnect service through Equinix
Just days after AT&T hooked up with IBM and Microsoft for cloud service connectivity, Verizon announced a software-defined interconnect (SDI) service to help the carrier's customers connect Equinix colocation data centers.These two companies have a history. In 2017, Equinix acquired 29 Verizon data centers in the U.S and Latin America for $3.6 billion. So like AT&T, Verizon left data centers to the experts and focused on building connections to them. [ Read also: How to plan a software-defined data center network ] As more enterprises move workloads into colocation facilities run by providers like Equinix, fast, secure connections between the enterprise and the colocation become a must. Verizon's SDI service is designed to provide fast, reliable connectivity between customer and the colocation data center.To read this article in full, please click here
0
Verizon launches interconnect service through Equinix
Just days after AT&T hooked up with IBM and Microsoft for cloud service connectivity, Verizon announced a software-defined interconnect (SDI) service to help the carrier's customers connect Equinix colocation data centers.These two companies have a history. In 2017, Equinix acquired 29 Verizon data centers in the U.S and Latin America for $3.6 billion. So like AT&T, Verizon left data centers to the experts and focused on building connections to them. [ Read also: How to plan a software-defined data center network ] As more enterprises move workloads into colocation facilities run by providers like Equinix, fast, secure connections between the enterprise and the colocation become a must. Verizon's SDI service is designed to provide fast, reliable connectivity between customer and the colocation data center.To read this article in full, please click here
0
Storage management a weak area for most enterprises
Stop me if you’ve heard this before: Companies are racing to a new technological paradigm but are using yesterday’s tech to do it.I know. Shocking.A survey of more than 300 storage professionals by storage vendor NGD Systems found only 11% of the companies they talked to would give themselves an “A” grade for their compute and storage capabilities.Why? The chief reason given is that while enterprises are rapidly deploying technologies for edge networks, real-time analytics, machine learning, and internet of things (IoT) projects, they are still using legacy storage solutions that are not designed for such data-intensive workloads. More than half — 54% — said their processing of edge applications is a bottleneck, and they want faster and more intelligent storage solutions.To read this article in full, please click here
0
Storage management a weak area for most enterprises
Stop me if you’ve heard this before: Companies are racing to a new technological paradigm but are using yesterday’s tech to do it.I know. Shocking.A survey of more than 300 storage professionals by storage vendor NGD Systems found only 11% of the companies they talked to would give themselves an “A” grade for their compute and storage capabilities.Why? The chief reason given is that while enterprises are rapidly deploying technologies for edge networks, real-time analytics, machine learning, and internet of things (IoT) projects, they are still using legacy storage solutions that are not designed for such data-intensive workloads. More than half — 54% — said their processing of edge applications is a bottleneck, and they want faster and more intelligent storage solutions.To read this article in full, please click here
0
IBM expands its storage management software to support competitor products
Most companies like to keep their specialty software to themselves, but IBM is making a move to expand its Spectrum Discover metadata management software to support competitive storage products.Spectrum Discover is modern metadata management software that provides data insight for petabyte-scale unstructured storage. It was designed for IBM Cloud Object Storage and IBM Spectrum Scale, a clustered file system, to rapidly ingest, consolidate and index metadata for billions of files and objects.Its metadata layer enables storage administrators, data stewards and data scientists to efficiently manage, classify, and gain insights from massive amounts of unstructured data while helping to mitigate risk and accelerate large-scale analytics.To read this article in full, please click here
0
IBM expands its storage management software to support competitor products
Most companies like to keep their specialty software to themselves, but IBM is making a move to expand its Spectrum Discover metadata management software to support competitive storage products.Spectrum Discover is modern metadata management software that provides data insight for petabyte-scale unstructured storage. It was designed for IBM Cloud Object Storage and IBM Spectrum Scale, a clustered file system, to rapidly ingest, consolidate and index metadata for billions of files and objects.Its metadata layer enables storage administrators, data stewards and data scientists to efficiently manage, classify, and gain insights from massive amounts of unstructured data while helping to mitigate risk and accelerate large-scale analytics.To read this article in full, please click here
0
AT&T partners with IBM and Microsoft, focuses on network capabilities
Earlier this year, AT&T finalized a deal to divest itself of its 31 data centers for $1.1 billion. Now that it has dumped its data center business, the company partnering with two of the largest providers of them: Microsoft and IBM.IBM and AT&T this week announced a multi-year strategic alliance where AT&T’s network and IBM Cloud will link up to provide software-defined network (SDN) services, including giving IBM Cloud access to AT&T’s 5G network. In return, IBM will make AT&T its primary provider of 5G, edge computing, and internet of things (IoT) services and help manage AT&T’s entire infrastructure footprint, including third-party cloud services, using Red Hat’s open-source tools to manage the network. This isn’t really new, as AT&T was using Red Hat Enterprise Linux (RHEL) for some time.To read this article in full, please click here