Intel looking likely to manufacture Nvidia chips

More than a year ago, Nvidia CEO Jensen Huang said he was open to the possibility of having Intel manufacture Nvidia’s GPUs through Intel's foundry services program.At the time, Huang was noncommittal beyond saying that Nvidia was looking at the possibility. Now things are getting more concrete. During a question-and-answer session at the Computex tradeshow in Taipei, Taiwan, Huang said he had recently received good results for an Intel test chip based on the company's next-generation process node."You know that we also manufacture with Samsung, and we're open to manufacturing with Intel. [Intel CEO Pat Gelsinger] has said in the past that we're evaluating the process, and we recently received the test chip results of their next-generation process, and the results look good," Huang said.To read this article in full, please click here

Writing An IETF Draft: Document Streams And Document Status

So far in this series we’ve discussed the history of the IETF, some of the tools you might want to use when building an IETF submission, and document formatting. There are other seemingly mystical concepts in the IETF process as well—for instance, what is a “document stream,” and what is a document’s “status?” Let’s look […]

The post Writing An IETF Draft: Document Streams And Document Status appeared first on Packet Pushers.

6 lessons from the Amazon Prime Video serverless vs. monolith flap

A software-development team caused quite a stir recently with a blog post describing how it abandoned a serverless architecture project in favor of a monolith—and slashed cloud infrastructure costs by 90% in the process.But this wasn’t just any team; the post was written by Marcin Kolny, a senior software-development engineer at Amazon Prime Video.Since Amazon is one of the leading advocates for serverless computing, not to mention the market leader in cloud services, the post was viewed as either a commendable act of openness or the very definition of throwing your company under the bus. Either way, it triggered a passionate back and forth on social media platforms that focused on larger questions:To read this article in full, please click here

Day Two Cloud 197: Understanding Decentralized Identity With Dr. Joanne Friedman

On today's Day Two Cloud we dive into Decentralized Identity, or DID. This W3C standard is a unique identifier, authenticated by cryptographic proofs, that individuals and organizations can generate themselves using systems they trust. A DID can be a person, an object, an organization, a data model, or other subject. We explore DIDs, how they work, and what they can be used for with Dr. Joanne Friedman.

Day Two Cloud 197: Understanding Decentralized Identity With Dr. Joanne Friedman

On today's Day Two Cloud we dive into Decentralized Identity, or DID. This W3C standard is a unique identifier, authenticated by cryptographic proofs, that individuals and organizations can generate themselves using systems they trust. A DID can be a person, an object, an organization, a data model, or other subject. We explore DIDs, how they work, and what they can be used for with Dr. Joanne Friedman.

The post Day Two Cloud 197: Understanding Decentralized Identity With Dr. Joanne Friedman appeared first on Packet Pushers.

NTP in a Nutshell

Years ago I’ve been involved in an interesting discussion focusing on NTP authentication and whether you can actually implement it reliably on Cisco IOS. What I got out of it (apart from a working example) was the feeling that NTP and it’s implementation in Cisco IOS was under-understood and under-documented, so I wrote an article about it. Of course the web version got lost in the mists of time but I keep my archives handy.

Last weekend I migrated that article to blog.ipSpace.net. I hope you’ll still find it useful; while it’s pretty old, the fundamentals haven’t changed in the meantime.

NTP in a Nutshell

Years ago I’ve been involved in an interesting discussion focusing on NTP authentication and whether you can actually implement it reliably on Cisco IOS. What I got out of it (apart from a working example) was the feeling that NTP and it’s implementation in Cisco IOS was under-understood and under-documented, so I wrote an article about it. Of course the web version got lost in the mists of time but I keep my archives handy.

Last weekend I migrated that article to blog.ipSpace.net. I hope you’ll still find it useful; while it’s pretty old, the fundamentals haven’t changed in the meantime.

Heavy Wireless 003: Designing Eduroam For The Future With Mark Houtz

On today's Heavy Wireless, Keith Parsons talks with guest is Mark Houtz about designing Eduroam for the future, including the 6Ghz band. Mark explains the challenges of filtering Internet access for K-12 students when they travel outside of their local high school and the technical requirements of Eduroam, specifically the use of WPA3 Enterprise on all frequencies when using the new 6 GHz band. Mark also shares his testing process for Eduroam's compatibility with 6 GHz radios.

Heavy Wireless 003: Designing Eduroam For The Future With Mark Houtz

On today's Heavy Wireless, Keith Parsons talks with guest is Mark Houtz about designing Eduroam for the future, including the 6Ghz band. Mark explains the challenges of filtering Internet access for K-12 students when they travel outside of their local high school and the technical requirements of Eduroam, specifically the use of WPA3 Enterprise on all frequencies when using the new 6 GHz band. Mark also shares his testing process for Eduroam's compatibility with 6 GHz radios.

The post Heavy Wireless 003: Designing Eduroam For The Future With Mark Houtz appeared first on Packet Pushers.

Network Break 432: DriveNets Aims To Make Ethernet AI-Friendly; China Goes Eye-For-An-Eye With US Over Tech Bans

This week we discuss a new offering from DriveNets that aims to make Ethernet more suitable as a network fabric for AI workloads, why the EU slapped Meta with a $1.3 billion fine, and a new offering from Extreme Networks that lets you manage Extreme gear in the cloud or from on-prem, China banning Micron, and more IT news.

The post Network Break 432: DriveNets Aims To Make Ethernet AI-Friendly; China Goes Eye-For-An-Eye With US Over Tech Bans appeared first on Packet Pushers.

Tech Bytes: Assembling A SASE Architecture With Fortinet (Sponsored)

Today on the Tech Bytes podcast we pull back the covers on SASE, or Secure Access Service Edge. Fortinet is our sponsor. One important concept to grasp around SASE is that it’s as much an architecture as it is a product. It requires planning and foresight to put the pieces together and operate them. We talk with Fortinet about the elements of its SASE offering and what a typical customer engagement with SASE looks like.

Tech Bytes: Assembling A SASE Architecture With Fortinet (Sponsored)

Today on the Tech Bytes podcast we pull back the covers on SASE, or Secure Access Service Edge. Fortinet is our sponsor. One important concept to grasp around SASE is that it’s as much an architecture as it is a product. It requires planning and foresight to put the pieces together and operate them. We talk with Fortinet about the elements of its SASE offering and what a typical customer engagement with SASE looks like.

The post Tech Bytes: Assembling A SASE Architecture With Fortinet (Sponsored) appeared first on Packet Pushers.

How doNotTrack policies work in the Calico eBPF dataplane

Almost all modern network systems, including stateful firewalls, make use of connection tracking (“conntrack”) because it consumes less processing power per packet and simplifies operations. However, there are use cases where connection tracking has a negative impact, as we described in Linux Conntrack: Why it breaks down and avoiding the problem. Distributed Denial of Service (DDoS) mitigation systems, defending against volumetric network attacks, is a well known example of such a use case, as it needs to drop malicious packets as fast as possible. In addition to these attacks, connection tracking becomes a potential attack vector as it is a limited resource. There are also applications generating huge amounts of short lived connections per second, to the point that tracking connections leads to more processing and defeating its intended purposes. These use cases demonstrate that there is a need to not track connections in a firewall, also known as stateless firewalling.

In this blog post, we will explain how Project Calico uses eXpress Data Path (XDP) in its eBPF dataplane (also in its iptables dataplane but not the focus of this post) to improve the performance of its stateless firewall. XDP is an eBPF hook that allows a program to Continue reading

MGX: Nvidia Standardizes Multi-Generation Server Designs

Whenever a compute engine maker also does motherboards as well as system designs, those companies that make motherboards (there are dozens who do) and create system designs (the original design manufacturers and the original – get a little bit nervous as well as a bit relieved. …

MGX: Nvidia Standardizes Multi-Generation Server Designs was written by Timothy Prickett Morgan at The Next Platform.

Qualcomm doubles down on its pivot to AI

Qualcomm has announced it is shifting its focus from providing chips exclusively for communications devices and doubling down on its efforts to support AI workloads.The company is transitioning to becoming  an “intelligent edge computing” firm, Alex Katouzian, a senior vice president at Qualcomm, said during a keynote speech at the Computex show in Taipei Tuesday.AI workloads require a lot of compute power and in February, Qualcomm announced the Snapdragon X75, its latest 5G modem component that the company said will be the world’s first modem-RF system for 5G-Advanced — a set of specifications designed to improve speed, maximize coverage, and enhance mobility and power efficiency for mobile devices. The X75 is also reportedly able to process AI workloads 2.5 times faster than its predecessor, the X70.To read this article in full, please click here

Qualcomm doubles down on its pivot to AI during Computex keynote

Qualcomm has announced it is shifting its focus from providing chips exclusively for communications devices and doubling down on its efforts to support AI workloads.The company is transitioning to becoming  an “intelligent edge computing” firm, Alex Katouzian, a senior vice president at Qualcomm, said during a keynote speech at the Computex show in Taipei Tuesday.AI workloads require a lot of compute power and in February, Qualcomm announced the Snapdragon X75, its latest 5G modem component that the company said will be the world’s first modem-RF system for 5G-Advanced — a set of specifications designed to improve speed, maximize coverage, and enhance mobility and power efficiency for mobile devices. The X75 is also reportedly able to process AI workloads 2.5 times faster than its predecessor, the X70.To read this article in full, please click here

Inside Nvidia’s new AI supercomputer

With Nvidia’s Arm-based Grace processor at its core, the company has introduced a supercomputer designed to perform AI processing powered by a CPU/GPU combination.The new system, formally introduced at the Computex tech conference in Taipei the DGX GH200 supercomputer is powered by 256 Grace Hopper Superchips, technology that is a combination of Nvidia’s Grace CPU, a 72-core Arm processor designed for high-performance computing and the Hopper GPU. The two are connected by Nvidia’s proprietary NVLink-C2C high-speed interconnect.To read this article in full, please click here

1 … 264 265 266 267 268 … 3,819