Ultra Ethernet: NSCC Destination Flow Control

Figure 6-14 depicts a demonstrative event where Rank 4 receives seven simultaneous flows (1). As these flows are processed by their respective PDCs and handed over to the Semantic Sublayer (2), the High-Bandwidth Memory (HBM) Controller becomes congested. Because HBM must arbitrate multiple fi_write RMA operations requiring concurrent memory bank access and state updates, the incoming packet rate quickly exceeds HBM’s transactional retirement rate. 

This causes internal buffers at the memory interface to fill, creating a local congestion event (3). To prevent buffer overflow, which would lead to dropped packets and expensive RMA retries, the receiver utilizes NSCC to move the queuing "pain" back to the source. This is achieved by using pds.rcv_cwnd_pend parameter of the ACK_CC header (4). The parameter operates on a scale of 0 to 127; while zero is ignored, a value of 127 triggers the maximum possible rate decrement. In this scenario, a value of 64 is utilized, resulting in a 50% penalty relative to the newly acknowledged data.

Rather than directly computing a new transport rate, the mechanism utilizes a three-phase process to define a restricted Congestion Window (CWND). This reduction in CWND inherently forces the source to drain its inflight bucket to Continue reading

Introducing the New Tigera & Calico Brand

Same community. A clearer, more unified look.

Today, we are excited to share a refresh of the Tigera and Calico visual identity!

This update better reflects who we are, who we serve, and where we are headed next.

If you have been part of the Calico community for a while, you know that change at Tigera is always driven by substance, not style alone. Since the early days of Project Calico, our focus has always been clear: Build powerful, scalable networking and security for Kubernetes, and do it in the open with the community.

Built for the Future, With the Community

Tigera was founded by the original Project Calico engineering team and remains deeply committed to maintaining Calico Open Source as the leading standard for container networking and network security.

“Tigera’s story began in 2016 with Project Calico, an open-source container networking and security project. Calico Open Source has since become the most widely adopted solution for containers and Kubernetes. We remain committed to maintaining Calico Open Source as the leading standard, while also delivering advanced capabilities through our commercial editions.”
—Ratan Tipirneni, President & CEO, Tigera

A Visual Evolution

This refresh is an evolution, not a reinvention. You Continue reading

TNO054: AI Skills for CCIEs

Let’s talk about AI for NetOps: It’s not just coming, it’s here. There are tools to use, skills to acquire, and we want to talk about what’s needed for highly certified network engineers to skill up in AI. What certification opportunities or paths exist? What developments do we think we’re going to see here? And... Read more »

Google’s AI advantage: why crawler separation is the only path to a fair Internet

Earlier this week, the UK’s Competition and Markets Authority (CMA) opened its consultation on a package of proposed conduct requirements for Google. The consultation invites comments on the proposed requirements before the CMA imposes any final measures. These new rules aim to address the lack of choice and transparency that publishers (broadly defined as “any party that makes content available on the web”) face over how Google uses search to fuel its generative AI services and features. These are the first consultations on conduct requirements launched under the digital markets competition regime in the UK. 

We welcome the CMA’s recognition that publishers need a fairer deal and believe the proposed rules are a step into the right direction. Publishers should be entitled to have access to tools that enable them to control the inclusion of their content in generative AI services, and AI companies should have a level playing field on which to compete. 

But we believe the CMA has not gone far enough and should do more to safeguard the UK’s creative sector and foster healthy competition in the market for generative and agentic AI. 

CMA designation of Google as having Strategic Market Status 

In January Continue reading

Building vertical microfrontends on Cloudflare’s platform

Updated at 6:55 a.m. PT

Today, we’re introducing a new Worker template for Vertical Microfrontends (VMFE). This template allows you to map multiple independent Cloudflare Workers to a single domain, enabling teams to work in complete silos — shipping marketing, docs, and dashboards independently — while presenting a single, seamless application to the user.

Most microfrontend architectures are "horizontal", meaning different parts of a single page are fetched from different services. Vertical microfrontends take a different approach by splitting the application by URL path. In this model, a team owning the `/blog` path doesn't just own a component; they own the entire vertical stack for that route – framework, library choice, CI/CD and more. Owning the entire stack of a path, or set of paths, allows teams to have true ownership of their work and ship with confidence.

Teams face problems as they grow, where different frameworks serve varying use cases. A marketing website could be better utilized with Astro, for example, while a dashboard might be better with React. Or say you have a monolithic code base where many teams ship as a collective. An update to add new features from several teams can get frustratingly rolled back because Continue reading

Ultra Ethernet: Inflight Bytes and CWND Adjustment

Inflight Packet Adjustment

Figure 6-12 depicts the ACK_CC header structure and fields. When NSCC is enabled in the UET node, the PDS must use the pds.type ACK_CC in the prologue header, which serves as the common header structure for all PDS messages. Within the actual PDS ACK_CC header, the pds.cc_type must be set to CC_NSCC. The pds.ack_cc_state field describes the values and states for service_time, rc (restore congestion CWND), rcv_cwnd_pend, and received_bytes. The source specifically utilizes the received_bytes parameter to calculate the updated state for inflight packets.

The CCC computes the reduction in the inflight state by subtracting the rcvd_bytes value received in previous ACK_CC messages from the rcvd_bytes value carried within the latest ACK_CC message. As illustrated in Figure 6-12, the inflight state is decreased by 4,096 bytes, which is the delta between 16,384 and 12,288 bytes.

Recap: In order to transport data to network, the Inflight bytes must be less than CWND size.



Figure 6-12: NSCC: Inflight Bytes adjustment.


CWND Adjustment


A single, shared Congestion Window (CWND) regulates the total volume of bytes across all PDCs that are permitted for transmission to the backend network. The transport rate and network performance are continuously monitored and Continue reading

Introducing Moltworker: a self-hosted personal AI agent, minus the minis

The Internet woke up this week to a flood of people buying Mac minis to run Moltbot (formerly Clawdbot), an open-source, self-hosted AI agent designed to act as a personal assistant. Moltbot runs in the background on a user's own hardware, has a sizable and growing list of integrations for chat applications, AI models, and other popular tools, and can be controlled remotely. Moltbot can help you with your finances, social media, organize your day — all through your favorite messaging app.

But what if you don’t want to buy new dedicated hardware? And what if you could still run your Moltbot efficiently and securely online? Meet Moltworker, a middleware Worker and adapted scripts that allows running Moltbot on Cloudflare's Sandbox SDK and our Developer Platform APIs.

A personal assistant on Cloudflare — how does that work? 

Firstly, Cloudflare Workers has never been so compatible with Node.js. Where in the past we had to mock APIs to get some packages running, now those APIs are supported natively by the Workers Runtime.

This has changed how we can build tools on Cloudflare Workers. When we first implemented Playwright, a popular framework for web testing and automation that runs Continue reading

Multicast PIM Dense Mode (III)

Multicast PIM Dense Mode (III)

In the previous posts, we covered the basics of multicast, including how sources send traffic to group addresses and how receivers use IGMP to signal their interest to the last hop router. We also looked at how IGMP snooping helps switches forward multicast traffic only to ports with interested receivers.

In this post, we will look at PIM, Protocol Independent Multicast. PIM is the protocol that routers use to build the multicast forwarding tree between the source and the receivers. There are different modes/flavours of PIM, and in this post, we will focus specifically on PIM Dense Mode.

Multicast Introduction
With multicast, the source sends only a single copy of the traffic into the network. As that traffic moves through the network, it is replicated
Multicast PIM Dense Mode (III)PacketswitchSuresh Vinasiththamby
Multicast PIM Dense Mode (III)
Multicast IGMP - Internet Group Management Protocol
IGMP is the protocol used by receivers to signal their interest in multicast traffic. When a host wants to receive a multicast stream, it sends
Multicast PIM Dense Mode (III)PacketswitchSuresh Vinasiththamby
Multicast PIM Dense Mode (III)

PIM Overview

PIM, Protocol Independent Multicast, is the protocol routers use to build a loop-free multicast distribution tree from the source to the receivers. It is called protocol-independent because it does not rely on any specific unicast Continue reading

Big Blue Poised To Peddle Lots Of On Premises GenAI

If you want to know the state of the art in GenAI model development, you watch what the Super 8 hyperscalers and cloud builders are doing and you also keep an eye on the major model builders outside of these companies – mainly, OpenAI, Anthropic, and xAI as well as a few players in China like DeepSeek. …

Big Blue Poised To Peddle Lots Of On Premises GenAI was written by Timothy Prickett Morgan at The Next Platform.

Why Kubernetes Flat Networks Fail at Scale—and Why Your Cluster Needs a Security Hierarchy

Kubernetes networking offers incredible power, but scaling that power often transforms a clean architecture into a tangled web of complexity. Managing traffic flow between hundreds of microservices across dozens of namespaces presents a challenge that touches every layer of the organization, from engineers debugging connections to the architects designing for compliance.

The solution to these diverging challenges lies in bringing structure and validation to standard Kubernetes networking. Here is a look at how Calico Tiers and Staged Network Policies help you get rid of this networking chaos.

The Limits of Flat Networking

The default Kubernetes NetworkPolicy resource operates in a flat hierarchy. In a small cluster, this is manageable. However, in an enterprise environment with multiple tenants, teams, and compliance requirements, “flat” quickly becomes unmanageable, and dangerous.

To make this easier, imagine a large office building where every single employee has a key that opens every door. To secure the CEO’s office in a flat network, you have to put “Do Not Enter” signs on every door that could lead to it. That is flat networking, secure by exclusion rather than inclusion.

Without a security hierarchy, every new policy risks becoming a potential mistake that overrides others, and debugging connectivity Continue reading

Worth Reading 012826


People often invest in companies not because they think the companies will succeed, but because they think others will invest. Thus, they try to invest before the others do, hoping there will be a greater fool who will pay more than they did.


More than half of CEOs report seeing neither increased revenue nor decreased costs from AI, despite massive investments in the technology, according to a PwC survey of 4,454 business leaders.


DNSSEC promised to secure DNS with cryptographic proof, yet messy rollouts, outages, and hype backlash ruined its reputation.


If you start the query process with an IP address, or an AS Number then you can find details about the current holder of that number resource, including its organisation name. But the reverse query is far more challenging in this database query framework.


Earlier this month, OpenAI introduced a new health focused space within ChatGPT, pitching it as a safer way for users to ask questions about sensitive topics like medical data, illnesses, and fitness.


Let’s start with hosting, and the key point here is that resource pools to host AI agents are not the answer. There are multiple reasons for this, but the main one is that Continue reading

1 … 13 14 15 16 17 … 3,856