Archive

Category Archives for "Networking"

Cisco Nexus data-center switches promise 800Gb Ethernet but deliver 400GbE today

Cisco is prepping its high-end data-center Nexus switch family for high-speed Ethernet  that will better support high-bandwidth applications such as AI and cloud-native.  The vendor rolled out five new Nexus data-center switches that include improved support for 100/400Gb Ethernet networks, but its flagship box—the 9800—now includes a migration path to support 800GbE requirements. [ Get regularly scheduled insights by signing up for Network World newsletters. ] The new modular data-center switch product family with 14.4Tb/s fully encrypted bandwidth per slot. The Cisco Nexus 9800 Series features eight-slot and four-slot chassis that can scale from 57Tb/s to 115Tb/s with a combination of various first-generation line cards and fabric modules. Each line card slot can support 400GbE or 100GbE or 10/25/50GbE ports. The eight-slot option could support up to 288 400GbE ports, well above the current 9000’s 32-port capacity.To read this article in full, please click here

Cloudflare Zaraz launches new privacy features in response to French CNIL standards

Cloudflare Zaraz launches new privacy features in response to French CNIL standards
Cloudflare Zaraz launches new privacy features in response to French CNIL standards

Last week, the French national data protection authority (the Commission Nationale de l'informatique et des Libertés or “CNIL”), published guidelines for what it considers to be a GDPR-compliant way of loading Google Analytics and similar marketing technology tools. The CNIL published these guidelines following notices that the CNIL and other data protection authorities issued to several organizations using Google Analytics stating that such use resulted in impermissible data transfers to the United States. Today, we are excited to announce a set of features and a practical step-by-step guide for using Zaraz that we believe will help organizations continue to use Google Analytics and similar tools in a way that will help protect end user privacy and avoid sending EU personal data to the United States. And the best part? It takes less than a minute.

Enter Cloudflare Zaraz.

The new Zaraz privacy features

What we are releasing today is a new set of privacy features to help our customers enhance end user privacy. Starting today, on the Zaraz dashboard, you can apply the following configurations:

  • Remove URL query parameters: when toggled-on, Zaraz will remove all query parameters from a URL that is reported to a third-party server. It will turn Continue reading

Cisco data-center switches promise 800Gb Ethernet, deliver 400GbE today

Cisco is prepping its high-end data-center Nexus switch family for high-speed Ethernet  that will better support high-bandwidth applications such as AI and cloud-native.  The vendor rolled out five new Nexus data-center switches that include improved support for 100/400Gb Ethernet networks, but its flagship box—the 9800—now includes a migration path to support 800GbE requirements. [ Get regularly scheduled insights by signing up for Network World newsletters. ] The new modular data-center switch product family with 14.4Tb/s fully encrypted bandwidth per slot. The Cisco Nexus 9800 Series features eight-slot and four-slot chassis that can scale from 57Tb/s to 115Tb/s with a combination of various first-generation line cards and fabric modules. Each line card slot can support 400GbE or 100GbE or 10/25/50GbE ports. The eight-slot option could support up to 288 400GbE ports, well above the current 9000’s 32-port capacity.To read this article in full, please click here

Cisco data-center switches promise 800Gb Ethernet, deliver 400GbE today

Cisco is prepping its high-end data-center Nexus switch family for high-speed Ethernet  that will better support high-bandwidth applications such as AI and cloud-native.  The vendor rolled out five new Nexus data-center switches that include improved support for 100/400Gb Ethernet networks, but its flagship box—the 9800—now includes a migration path to support 800GbE requirements. [ Get regularly scheduled insights by signing up for Network World newsletters. ] The new modular data-center switch product family with 14.4Tb/s fully encrypted bandwidth per slot. The Cisco Nexus 9800 Series features eight-slot and four-slot chassis that can scale from 57Tb/s to 115Tb/s with a combination of various first-generation line cards and fabric modules. Each line card slot can support 400GbE or 100GbE or 10/25/50GbE ports. The eight-slot option could support up to 288 400GbE ports, well above the current 9000’s 32-port capacity.To read this article in full, please click here

Cisco Live 2022: A Kinder, Gentler, Cloudier Monster?

Cisco Live 2022 in Las Vegas kicked off with executive keynotes, including an address from CEO Chuck Robbins. My takeaways from the keynotes from Tuesday, June 14th are: Cisco knows it has to work harder to keep customers Cisco has big cloud ambitions Meraki is one key to Cisco’s cloud & simplicity goals Cisco Has […]

The post Cisco Live 2022: A Kinder, Gentler, Cloudier Monster? appeared first on Packet Pushers.

Hedge 134: Ten Things

One of the many reasons engineers should work for a vendor, consulting company, or someone other than a single network operator at some point in their career is to develop a larger view of network operations. What are common ways of doing things? What are uncommon ways? In what ways is every network broken? Over time, if you see enough networks, you start seeing common themes and ideas. Just like history, networks might not always be the same, but the problems we all encounter often rhyme. Ken Calenza joins Tom Ammon, Eyvonne Sharp, and Russ White to discuss these common traits—ten things I know about your network.

download

10 top automation and orchestration tools

IT process automation sells itself: automating tasks is not only cheaper than paying a human to perform repetitive activities, but it’s also more efficient and predictable.While it’s possible to develop automation tools in-house with enterprise staff, that can be challenging, so ultimately, to embrace automation in a big way, it may be necessary to enlist commercial software tools. [ Get regularly scheduled insights by signing up for Network World newsletters. ] Some tasks are simpler to automate than others—managing IT systems, provisioning physical and virtual machines, managing server configuration, identifying policy drift—and many IT systems are now built with features that make it easier to move along the path toward automation without seeking commercial platforms.To read this article in full, please click here

EVPN-VXLAN Explainer 5 – Layer 3 with Asymmetrical IRB

EVPN-VXLAN Explainer 5 - Layer 3 with Asymmetrical IRB

Thus far, this series of posts have all been about Layer 2 over Layer 3 models; the customer ethernet frames encapsulated in UDP, traversing L3 networks. The routing has been confined underlay, the customer traffic has stayed within the same network.
No longer! In this post, things start getting a little more interesting, as we look at routing the customer traffic with an EVPN feature called Integrated Routing and Bridging, or IRB.

  • First we look at the concept of routing in VXLAN networks.
  • Then we have an in-depth look at asymmetrical IRB (I'll be dealing with symmetrical in the next post).

✅ L2 is intra-subnet, L3 is inter-subnet

📥 Intra-subnet

To define terms, when I say 'intra-subnet', that is L2 traffic transferred between nodes in the same subnet.

📤 Inter-subnet

'Inter-subnet' refers to a traffic flow that traverses subnet boundaries.

☎️ The Centralized IP L3 Gateways of Old

  • With VXLAN networks in the past, inter-subnet communication was often performed by a centralized, IP only, gateway on behalf of the rest of the network.
  • Traffic from customer-side networks would need to be sent to this central device for routing, which often created inefficient traffic flows, and possibly a bandwidth choke-point.
  • Imagine Continue reading

Digital infrastructure outages get more costly

Digital infrastructure outages have gotten more and more expensive over the course of the past several years, according to a report from the Uptime Institute. Meanwhile, the total number of major outages has remained the same—meaning that, on average, an increasingly large amount of money is getting spent on recovering from each disruption.The proportion of individual outages resulting in losses of over $100,000 is increasing, according to the report, up to 47% of all outages in 2021 from 40% in the previous year. The institute said that, while it doesn't calculate an average overall cost per outage, overall trends are toward more costs being incurred by the average outage.To read this article in full, please click here

AMD ups its supercomputer components

AMD is working on an accelerated processing unit that will outperform its current top APU that powers the world’s first exascale supercomputer.At its recent analyst day, the company introduced a new high-end accelerator, the Instinct MI300, an APU that combines Zen 4 CPUs, the latest generation of GPU technology, plus AMD’s Infinity Cache and Infinity architecture in one package. It will deliver eight times the AI performance of AMD’s current high-end ACU, the MI250, and will be available next year.A pool of high-bandwidth memory on the ACU is shared between the CPU and the GPU allowing them to communicate freely without the performance or energy overhead of redundant memory copies.To read this article in full, please click here

AMD ups its supercomputer components

AMD is working on an accelerated processing unit that will outperform its current top APU that powers the world’s first exascale supercomputer.At its recent analyst day, the company introduced a new high-end accelerator, the Instinct MI300, an APU that combines Zen 4 CPUs, the latest generation of GPU technology, plus AMD’s Infinity Cache and Infinity architecture in one package. It will deliver eight times the AI performance of AMD’s current high-end ACU, the MI250, and will be available next year.A pool of high-bandwidth memory on the ACU is shared between the CPU and the GPU allowing them to communicate freely without the performance or energy overhead of redundant memory copies.To read this article in full, please click here

Securing cloud workloads in 5 easy steps

As organizations transition from monolithic services in traditional data centers to microservices architecture in a public cloud, security becomes a bottleneck and causes delays in achieving business goals. Traditional security paradigms based on perimeter-driven firewalls do not scale for communication between workloads within the cluster and 3rd-party APIs outside the cluster. The traditional paradigm also does not provide granular access controls to the workloads and zero-trust architecture, leaving cloud-native applications with a larger attack surface.

Calico Cloud offers an easy 5-step process for fast-tracking your organization’s cloud-native application journey by making security a business enabler while mitigating risk.

Step 1: Visibility

Gaining visibility into workload-to-workload communication with all metadata context intact is one of the biggest challenges when it comes to deploying microservices. You can’t apply security controls to what you can’t see. The traffic is not just flowing from a client to a server in this new cloud native distributed architecture but also between namespaces that reside between many nodes, causing flow proliferation. With Calico Cloud, you get a dynamic visualization of all traffic flowing through your network in an easy-to-read UI.

Example 1: You can view all the inside and outside (east-west and north-south) connections directly from Calico’s Continue reading

Cisco service predicts SD-WAN problems

Cisco is set to offer the first fruits of its technology that promises to let enterprises proactively avoid network problems and increase application performance.At its Cisco Live event this week the company took the wraps off ThousandEyes WAN Insights service that will let Cisco SD-WAN customers get network forecasts and SD-WAN policy recommendations for elevating application performance and user experience across enterprise sites.[ Get daily insights by signing up for Network World newsletters. ] “Today’s hybrid work environments are incredibly complex, made up of highly sophisticated applications that are distributed across heterogeneous networks and accessed by end users from a variety of locations that may have vastly different underlying network conditions,” Mike Hicks, principal solutions analyst with Cisco ThousandEyes, wrote in a blog. "ThousandEyes WAN Insights helps IT operations teams anticipate changes to these environments using data-driven analysis that gives them actionable network recommendations to reduce issues and optimize performance across Internet, cloud, and SaaS."To read this article in full, please click here

Heavy Networking 635: Unified Network Fabrics With Juniper Apstra (Sponsored)

In today’s sponsored Heavy Networking we talk to Juniper Apstra about how how Apstra delivers on unified data center operations, why fabrics are everywhere, how Apstra differs from other automation and intent solutions, and more. Our guest is Mansour Karam, VP of Product Management.

The post Heavy Networking 635: Unified Network Fabrics With Juniper Apstra (Sponsored) appeared first on Packet Pushers.

Cisco moves Catalyst, Nexus management to the cloud

Cisco is taking a big step toward cloud-management of both its Catalyst campus and Nexus data-center equipment.At the Cisco Live customer event this week, the company rolled out two cloud-based management services that provide more options for enterprises to support hybrid workforces. [ Get regularly scheduled insights by signing up for Network World newsletters. ] Catalyst management in the cloud The first service, Cloud Management for Cisco Catalyst, lets customers manage and troubleshoot Catalyst 9000 switching and wireless campus and branch devices from the company’s cloud-based Meraki dashboard, which can manage and troubleshoot a wide variety of devices and networks from a single screen. According to Cisco, Catalyst customers can run a CLI command with information about their organization, and it will move management of that device over to the Meraki cloud.To read this article in full, please click here