--- title: "Lossless Ethernet for AI Fabrics: PFC, LLR or CBFC — Which to Choose and How to Verify It" description: "PFC+DCQCN, Link Layer Retry or CBFC: we compare lossless Ethernet options for AI fabrics, the pitfalls of each, and how to verify them on a test bed." image: https://test4net.com/hubfs/blog-covers/test4net-news-04.png --- [Skip to content](https://test4net.com/blog/lossless-ethernet-ai-fabrics-pfc-llr-cbfc#main-content) [![TEST4NET](https://test4net.com/hubfs/test4net-logo-fixed-2.svg)](https://test4net.com/?hsLang=en) - CLOUDLAB - SERVICES - SOLUTIONS - VENDORS - RESOURCES - COMPANY - [English](https://test4net.com/blog/lossless-ethernet-ai-fabrics-pfc-llr-cbfc) - [Русский](https://test4net.com/ru/blog/lossless-ethernet-ai-fabrics-pfc-llr-cbfc) - [Հայերեն](https://test4net.com/hy/blog/lossless-ethernet-ai-fabrics-pfc-llr-cbfc) [Login](https://lab.test4net.com) [Support](https://test4net.com/support?hsLang=en) - [English](https://test4net.com/blog/lossless-ethernet-ai-fabrics-pfc-llr-cbfc) - [Русский](https://test4net.com/ru/blog/lossless-ethernet-ai-fabrics-pfc-llr-cbfc) - [Հայերեն](https://test4net.com/hy/blog/lossless-ethernet-ai-fabrics-pfc-llr-cbfc) How It Works Remote access to a live test lab: pick a scenario, run it, get the report Equipment & Capabilities Traffic generators, emulators and DUT racks available in the lab Access Plans Hourly sessions, subscriptions and dedicated racks [Log in to CloudLAB POPULAR Your sessions, reservations and reports](https://lab.test4net.com) [Open the CloudLAB portal](https://lab.test4net.com) Network Test & Validation Load & Performance Testing Functional Testing Conformance Testing Interoperability Testing Security & Resilience Testing All testing services Automation & Expertise Test Automation & CI/CD Test Plan & Methodology Design Training Programs Lab & Equipment Lab Rental Hosted DUT Testing Equipment Repair Not sure where to start? Get a test plan for your setup Tell us what you run and what worries you. We propose the tests, tools and timeline. [Request a test plan](https://test4net.com/contacts?hsLang=en) By Technology By Industry By Task Routing & Switching Firewalls & NGFW SD-WAN & WAN Cloud & NFV 5G & Mobile Core Wi-Fi Storage & Infrastructure AI / Switching Fabric (RoCEv2) All technologies Telecom Operators Network & Security Vendors Banks & Financial Services System Integrators Enterprise IT Teams Managed Service Providers Media, Broadcast & Streaming Government & Public Sector All industries Pre-deployment Validation Vendor Selection Cloud Migration Readiness AI/GPU Cluster & RoCEv2 Readiness New Technology Rollout Troubleshooting & Benchmarking Disaster Recovery & Failover Testing SLA Validation & Vendor Accountability All tasks Testing Tools & Lab Ixia / Keysight POPULAR Emblasoft NEW Apposite Technologies T-Plan Cubro Candela Technologies Netitest How we integrate the tools IT Infrastructure HPE Vinchin Yeastar Become a partner Read Technologies / Protocols Methodologies [Blog Case studies, test insights and lab notes](https://test4net.com/blog?hsLang=en) [Industry News What is changing in network testing and infrastructure](https://test4net.com/news?hsLang=en) FAQ How the lab, sessions and reports work BGP Route exchange, policy and convergence OSPF & IS-IS IGP adjacency, flooding and SPF timing MPLS & Segment Routing LDP, RSVP-TE and SR-MPLS label paths EVPN / VXLAN Overlay fabrics, MAC/IP routes, multi-homing IPsec & IKEv2 Tunnel setup, rekeying and throughput TLS / HTTPS Handshakes, cipher suites and inspection RoCEv2 Lossless Ethernet, PFC and ECN for AI fabrics Wi-Fi 6 / 6E / 7 802.11ax/be, OFDMA, MLO and roaming All protocols [RFC 2544 Throughput, latency, frame loss and back-to-back](https://test4net.com/methodologies/rfc-2544?hsLang=en) RFC 2889 LAN switch benchmarking: forwarding, MAC learning, congestion [RFC 9411 / RFC 3511 Firewall and NGFW performance benchmarking](https://test4net.com/methodologies/rfc-9411-rfc-3511?hsLang=en) RFC 6349 TCP throughput and buffer delay testing RFC 5180 IPv6 benchmarking methodology ITU-T Y.1564 Ethernet service activation: CIR, EIR, KPIs per service ITU-T Y.1731 Ethernet OAM: delay, jitter, loss and SLA monitoring TR-398 Wi-Fi in-premises performance test plan All methodologies From the blog [![Wi-Fi testing](https://test4net.com/hubfs/blog-covers/menu/menu-wifi-640.webp) Wi-Fi Wi-Fi Testing for High-Performance Use Cases: Why a Walk Test Is Not a Test](https://test4net.com/blog/wi-fi-testing-for-high-performance-use-cases-why-a-walk-test-is-not-a-test?hsLang=en) [![Security validation](https://test4net.com/hubfs/blog-covers/menu/menu-security-640.webp) Security One Test Is Not Enough: Why Security Infrastructure Needs Continuous Validation](https://test4net.com/blog/security-platform-continuous-validation-updates?hsLang=en) [![Security product testing](https://test4net.com/hubfs/blog-covers/menu/menu-lab-640.webp) Methodology How to Run a Security Product Test You Can Actually Trust](https://test4net.com/blog/how-to-run-a-security-product-test-you-can-actually-trust?hsLang=en) [About Us Independent lab in Armenia, team and history](https://test4net.com/about?hsLang=en) [Why TEST4NET What you get that vendor slides never show](https://test4net.com/why-test4net?hsLang=en) Our Lab & Certifications Equipment, methods and the standards we test against Partner Program Resell, integrate and co-test with us [Contacts Talk to an engineer, not a sales bot](https://test4net.com/contacts?hsLang=en) [Support portal](https://test4net.com/support?hsLang=en) - CLOUDLAB How It Works Equipment & Capabilities Access Plans [Log in to CloudLAB POPULAR](https://lab.test4net.com) - SERVICES Network Test & Validation Load & Performance Testing Functional Testing Conformance Testing Interoperability Testing Security & Resilience Testing All testing services Automation & Expertise Test Automation & CI/CD Test Plan & Methodology Design Training Programs Lab & Equipment Lab Rental Hosted DUT Testing Equipment Repair [Not sure where to start? Get a test plan for your setup](https://test4net.com/contacts?hsLang=en) - SOLUTIONS By Technology Routing & Switching Firewalls & NGFW SD-WAN & WAN Cloud & NFV 5G & Mobile Core Wi-Fi Storage & Infrastructure AI / Switching Fabric (RoCEv2) All technologies By Industry Telecom Operators Network & Security Vendors Banks & Financial Services System Integrators Enterprise IT Teams Managed Service Providers Media, Broadcast & Streaming Government & Public Sector All industries By Task Pre-deployment Validation Vendor Selection Cloud Migration Readiness AI/GPU Cluster & RoCEv2 Readiness New Technology Rollout Troubleshooting & Benchmarking Disaster Recovery & Failover Testing SLA Validation & Vendor Accountability All tasks - VENDORS Testing Tools & Lab Ixia / Keysight POPULAR Emblasoft NEW Apposite Technologies T-Plan Cubro Candela Technologies Netitest How we integrate the tools IT Infrastructure HPE Vinchin Yeastar Become a partner - RESOURCES Read [Blog](https://test4net.com/blog?hsLang=en) [Industry News](https://test4net.com/news?hsLang=en) FAQ Technologies / Protocols BGP OSPF & IS-IS MPLS & Segment Routing EVPN / VXLAN IPsec & IKEv2 TLS / HTTPS RoCEv2 Wi-Fi 6 / 6E / 7 All protocols Methodologies [RFC 2544](https://test4net.com/methodologies/rfc-2544?hsLang=en) RFC 2889 [RFC 9411 / RFC 3511](https://test4net.com/methodologies/rfc-9411-rfc-3511?hsLang=en) RFC 6349 RFC 5180 ITU-T Y.1564 ITU-T Y.1731 TR-398 All methodologies From the blog [Wi-Fi Testing for High-Performance Use Cases: Why a Walk Test Is Not a Test Wi-Fi](https://test4net.com/blog/wi-fi-testing-for-high-performance-use-cases-why-a-walk-test-is-not-a-test?hsLang=en) [One Test Is Not Enough: Why Security Infrastructure Needs Continuous Validation Security](https://test4net.com/blog/security-platform-continuous-validation-updates?hsLang=en) [How to Run a Security Product Test You Can Actually Trust Methodology](https://test4net.com/blog/how-to-run-a-security-product-test-you-can-actually-trust?hsLang=en) - COMPANY [About Us](https://test4net.com/about?hsLang=en) [Why TEST4NET](https://test4net.com/why-test4net?hsLang=en) Our Lab & Certifications Partner Program [Contacts](https://test4net.com/contacts?hsLang=en) [Login](https://lab.test4net.com) [Support](https://test4net.com/support?hsLang=en) [All news](https://test4net.com/news?hsLang=en) [News](https://test4net.com/blog/tag/news) # Lossless Ethernet for AI Fabrics: PFC, LLR or CBFC — Which to Choose and How to Verify It ![](https://test4net.com/hubfs/IMG_0070.jpg) [Alexander Zemskov](https://test4net.com/blog/author/az) October 19, 2026 5 min read ![Lossless Ethernet for AI Fabrics: PFC, LLR or CBFC — Which to Choose and How to Verify It — Photo: Yuriy Vertikov / Unsplash](https://test4net.com/hs-fs/hubfs/blog-covers/test4net-news-04.png?width=1600&height=840&name=test4net-news-04.png) In this article At the OCP Global Summit 2026 (San Jose, October 12–15), Keysight showed two interop demos: Link Layer Retry (LLR) at 1.6T between the INPT-1600GE tester and a switch based on Broadcom Tomahawk 6 (200G SerDes), and LLR plus Credit-Based Flow Control (CBFC) at 400GE with an Alibaba Cloud scale-up switch built on Tomahawk Ultra (100G SerDes). The main takeaway: an AI-fabric engineer now has not one way to build a "lossless network" but three. And the choice has to be made on measurements, not on vendor presentations. ## Why lossless at all RoCEv2 inherited its transport from InfiniBand: lossless, strictly in-order delivery. Any drop means a retransmit and a longer job completion time. In scale-up networks (XPU↔XPU inside a rack) the latency budget is measured in hundreds of nanoseconds: Tomahawk Ultra claims 250 ns of switch latency. ## The three options: how they work **PFC + ECN/DCQCN.** The established RoCEv2 approach. The switch marks ECN as the queue grows, the receiver sends a CNP, and the NIC throttles that specific queue. PFC is the safety net: when a buffer is nearly full, the port sends PAUSE to its neighbor and halts the entire traffic class. Keysight's blog describes PFC as a blunt instrument: it cannot identify the offending flow and suppresses neighboring flows along with it. **Link Layer Retry.** An option in the Ultra Ethernet 1.0 specification. The transmitter keeps frames in a replay buffer, numbers them via the preamble, and the receiver acknowledges with ACK/NACK carried in PCS control ordered sets. On a loss, it retransmits using a go-back-N scheme. The timeout is tied to the link RTT, about 1 µs. LLR does not address congestion; it corrects bit errors that FEC fails to catch. At 200G SerDes and 1.6T, such errors are no longer rare. **Credit-Based Flow Control.** Also part of UE 1.0; it can replace PFC or run alongside it. The sender transmits only when it holds credits for the receiver's buffer, and credits are returned as buffer space is freed. Accounting is per virtual channel: up to 32 VCs versus PFC's 8 classes, with 20-bit counters. The buffer is sized precisely per channel rather than by the "RTT + MTU per port, plus margin" formula. ## Comparison | Criterion | PFC + ECN/DCQCN | Link Layer Retry | CBFC | | --- | --- | --- | --- | | Protects against | Buffer overflow | Link errors (FEC/FCS) | Buffer overflow | | Where it operates | Hop-by-hop PAUSE + end-to-end ECN | A single link | A single link, per VC | | Granularity | 8 classes | Frame | Up to 32 VCs | | Buffer requirement | Headroom = rate × RTT, per port | Replay buffer per link | Fixed per VC, smaller than PFC | | Hardware | Any DC switch and RoCE NIC | ASIC + NIC/XPU with UE link layer (TH6, Tomahawk Ultra) | Same | | Interop status | Mature, but requires careful tuning | First public demos (OFC and OCP 2026) | First demos at 400/800GE | | Main trap | Head-of-line blocking, deadlock, PAUSE storm | Latency tail on retries, window size | Credit leak/desynchronization | ## Traps you only see on a test bed **PFC.** As early as 2016, Microsoft documented PFC deadlocks in production and PAUSE-frame storms from NICs. Headroom grows with distance: Keysight gives the example of 400GbE over 2 km with a 20 µs RTT — about 1 MB per port. On top of that, ECMP collisions cut effective bandwidth to roughly 60%. **LLR.** Go-back-N resends not only the lost frame but every frame transmitted after it. At high BER the p99.9 tail grows even with zero drops. The replay buffer must be verified under load, along with the behavior when the NACK itself is lost. **CBFC.** Credits are shared state on both ends of the link. Typical failure modes: desynchronization after a link failure, a wrong packet size in the accounting, a conflict with PFC on the same port. The IEEE 802.1 working group has formally asked UEC how PFC and CBFC are meant to coexist. ## How to test 1. **Conformance.** LLR/CBFC message exchange per the specification, negotiation via LLDP, counter boundary values. 2. **Interop.** Tester against another vendor's ASIC — the same format Keysight used in its demos: INPT-1600GE against TH6, then 400GE against Tomahawk Ultra. 3. **Error injection.** Artificial BER on the link, counting LLR retries, measuring p50/p99/p99.9 latency, not just the average. 4. **Incast and all-to-all.** For PFC/DCQCN — PAUSE frequency, buffer fill, ECN loop reaction time. For CBFC — credit utilization per VC. 5. **Deadlock scenarios.** Cyclic buffer dependencies; the pass criterion is recovery without manual intervention. ## What this means for the business Three mechanisms, three cost levels. PFC/DCQCN runs on today's hardware but requires lengthy tuning and carries the risk of cluster downtime from deadlock. LLR and CBFC promise predictability but require new ASICs (Tomahawk 6 at 102.4 Tbit/s, Tomahawk Ultra at 51.2 Tbit/s) and NICs/XPUs with UE support. The main risk is buying hardware from two vendors whose LLR implementations are both "per spec" yet fail to interoperate in practice: the first public interop demos took place only in March and October 2026. An hour of GPU-cluster downtime costs more than a week of testing. ## What to do Do not choose a mechanism based on presentations. Put the candidates on a test bed, run conformance, interop, error injection and incast, and compare the latency tails. If you are choosing a lossless Ethernet mechanism, contact TEST4NET: we will discuss your scenario and propose a test program. ## Sources - [Keysight Showcases Link Layer Retry Interoperability with Alibaba Cloud at 2026 OCP Global Summit (Business Wire)](https://www.businesswire.com/news/home/20261012704598/en/Keysight-Showcases-Link-Layer-Retry-Interoperability-with-Alibaba-Cloud-at-2026-OCP-Global-Summit) - [Keysight: LLR and CBFC interop with Broadcom Tomahawk Ultra at 800GE, OFC 2026 (Business Wire)](https://www.businesswire.com/news/home/20260316488886/en) - [Ultra Ethernet's Design Principles and Architectural Innovations (arXiv 2508.08906)](https://arxiv.org/pdf/2508.08906) - [IEEE 802.1 liaison to UEC, July 2025](https://www.ieee802.org/1/files/public/docs2025/liaison-UEC-coordination-0725.pdf) - [Keysight blog: Congestion Control in AI Networks](https://www.keysight.com/blogs/en/tech/traf-gen/2026/06/03/congestion-control-in-ai-network) - [Broadcom Ships Tomahawk Ultra (Converge Digest)](https://convergedigest.com/broadcom-ships-tomahawk-ultra-to-power-ai-scale-up/) - [Broadcom Ships Tomahawk 6 (StorageReview)](https://www.storagereview.com/news/broadcom-ships-tomahawk-6-switch-series-unleashing-102-4-tbps-switching-capacity) - [Microsoft Research: RDMA for Cloud Computing](https://www.microsoft.com/en-us/research/project/rdma-for-cloud-computing/) Share Pass it on to someone who runs a network Independent lab data is most useful in the hands of the people who make the call. Send this piece to a colleague. [LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Ftest4net.com%2Fblog%2Flossless-ethernet-ai-fabrics-pfc-llr-cbfc) [X](https://twitter.com/intent/tweet?url=https%3A%2F%2Ftest4net.com%2Fblog%2Flossless-ethernet-ai-fabrics-pfc-llr-cbfc&text=Lossless+Ethernet+for+AI+Fabrics%3A+PFC%2C+LLR+or+CBFC+%E2%80%94+Which+to+Choose+and+How+to+Verify+It) [Telegram](https://t.me/share/url?url=https%3A%2F%2Ftest4net.com%2Fblog%2Flossless-ethernet-ai-fabrics-pfc-llr-cbfc&text=Lossless+Ethernet+for+AI+Fabrics%3A+PFC%2C+LLR+or+CBFC+%E2%80%94+Which+to+Choose+and+How+to+Verify+It) [WhatsApp](https://wa.me/?text=Lossless+Ethernet+for+AI+Fabrics%3A+PFC%2C+LLR+or+CBFC+%E2%80%94+Which+to+Choose+and+How+to+Verify+It%20https%3A%2F%2Ftest4net.com%2Fblog%2Flossless-ethernet-ai-fabrics-pfc-llr-cbfc) Copy link More apps ![](https://test4net.com/hubfs/IMG_0070.jpg) Written by ### [Alexander Zemskov](https://test4net.com/blog/author/az) Alexander Zemskov leads network and security testing at TEST4NET, an independent, vendor-neutral testing lab in Yerevan, Armenia. With 19+ years in performance and load testing on IXIA, Spirent, Keysight, Xena and Xinertel platforms, he runs NGFW, WAF, IPS, DLP and SASE evaluations for telecom operators, banks and system integrators. [LinkedIn](https://www.linkedin.com/in/azemskov/) [Website](https://test4net.com/?hsLang=en) [Previous article Meta at OCP 2026: No AI Fabric Gets Accepted Without Benchmarks on 1000+ GPUs](https://test4net.com/blog/meta-ocp-2026-ai-fabric-validation-benchmarks?hsLang=en) ### Comments In this article Put it to the test ### Have a device to put through this? Send us the shortlist and the KPIs that decide it. We will come back with a scoped test plan, a timeline and a fixed price. [Talk to a test engineer](https://test4net.com/contacts?hsLang=en) Industry News ## More news [![Meta at OCP 2026: No AI Fabric Gets Accepted Without Benchmarks on 1000+ GPUs — Photo: İsmail Enes Ayhan / Unsplash](https://test4net.com/hubfs/blog-covers/test4net-news-03.png)](https://test4net.com/blog/meta-ocp-2026-ai-fabric-validation-benchmarks?hsLang=en) [News](https://test4net.com/blog/tag/news) ### [Meta at OCP 2026: No AI Fabric Gets Accepted Without Benchmarks on 1000+ GPUs](https://test4net.com/blog/meta-ocp-2026-ai-fabric-validation-benchmarks?hsLang=en) The key takeaway from Meta's post for OCP Global Summit 2026: the hyperscaler does not bring a single segment of its AI network online without a test run. Meta hands its … ![](https://test4net.com/hubfs/IMG_0070.jpg) Alexander Zemskov · Oct 18, 2026 · 5 min read [![Ultra Ethernet 1.1 vs RoCEv2: What Changes in the AI Fabric Transport and How to Test It — Photo: Yuriy Vertikov / Unsplash](https://test4net.com/hubfs/blog-covers/test4net-news-02.png)](https://test4net.com/blog/ultra-ethernet-1-1-vs-rocev2-ai-fabric-transport-testing?hsLang=en) [News](https://test4net.com/blog/tag/news) ### [Ultra Ethernet 1.1 vs RoCEv2: What Changes in the AI Fabric Transport and How to Test It](https://test4net.com/blog/ultra-ethernet-1-1-vs-rocev2-ai-fabric-transport-testing?hsLang=en) The bottom line On October 12, 2026, the Ultra Ethernet Consortium published specification 1.1. Sixteen months after version 1.0 (June 11, 2025), the standard moved beyon… ![](https://test4net.com/hubfs/IMG_0070.jpg) Alexander Zemskov · Oct 17, 2026 · 5 min read [![Keysight Unveils 1.6T Testers. What It Changes for Teams Running 400G — Photo: Kirill Sh / Unsplash](https://test4net.com/hubfs/blog-covers/test4net-news-01.png)](https://test4net.com/blog/keysight-1-6t-testers-what-changes-for-400g-teams?hsLang=en) [News](https://test4net.com/blog/tag/news) ### [Keysight Unveils 1.6T Testers. What It Changes for Teams Running 400G](https://test4net.com/blog/keysight-1-6t-testers-what-changes-for-400g-teams?hsLang=en) On October 13, 2026, Keysight announced two 1.6T Ethernet testers, the SERT 1600GE and the AresONE Plus 1600GE, and demonstrated them at the OCP Global Summit in San Jose… ![](https://test4net.com/hubfs/IMG_0070.jpg) Alexander Zemskov · Oct 16, 2026 · 5 min read [![TEST4NET LLC](https://test4net.com/hubfs/test4net-logo-fixed.svg)](https://test4net.com?hsLang=en) Independent testing and validation for telecom, information security and cloud infrastructure — real insights, not assumptions. [mailto:sales@test4net.com](mailto:sales@test4net.com) #### Navigate - [Home](https://test4net.com/?hsLang=en) - [Why TEST4NET](https://test4net.com/why-test4net?hsLang=en) - [About us](https://test4net.com/about?hsLang=en) - [Contact](https://test4net.com/contacts?hsLang=en) - [Blog](https://test4net.com/blog?hsLang=en) - [Industry News](https://test4net.com/news?hsLang=en) #### Get in touch Sales [sales@test4net.com](mailto:sales@test4net.com) Support [support.test4net.com](https://support.test4net.com) © 2026 TEST4NET LLC. All rights reserved. [Privacy Policy](https://test4net.com/privacy-policy?hsLang=en) [Legal](https://test4net.com/legal?hsLang=en) ```json { "@context" : "https://schema.org", "@type" : "BlogPosting", "author" : { "@type" : "Person", "name" : "Alexander Zemskov", "url" : "https://test4net.com/blog/author/az" }, "dateModified" : "2026-10-19T06:00:02.372Z", "datePublished" : "2026-10-19T06:00:00.000Z", "headline" : "Lossless Ethernet for AI Fabrics: PFC, LLR or CBFC — Which to Choose and How to Verify It", "image" : [ "https://test4net.com/hubfs/blog-covers/test4net-news-04.png" ], "mainEntityOfPage" : { "@id" : "https://test4net.com/blog/lossless-ethernet-ai-fabrics-pfc-llr-cbfc", "@type" : "WebPage" }, "publisher" : { "@type" : "Organization", "logo" : { "@type" : "ImageObject", "url" : "https://test4net.com/hubfs/test4net-logo-fixed.svg" }, "name" : "TEST4NET LLC" } } ``` ```json { "@context" : "https://schema.org", "@id" : "https://test4net.com/#organization", "@type" : "Organization", "address" : { "@type" : "PostalAddress", "addressCountry" : "AM", "addressLocality" : "Yerevan", "postalCode" : "0033", "streetAddress" : "3 Hakob Hakobyan Street (Mergelyan Hub)" }, "alternateName" : "TEST4NET", "contactPoint" : { "@type" : "ContactPoint", "availableLanguage" : [ "en", "ru", "hy" ], "contactType" : "sales", "email" : "sales@test4net.com", "telephone" : "+374-60-88-88-99" }, "description" : "Independent, vendor-neutral lab for testing network and cybersecurity products — NGFW, WAF, IPS, DLP and SASE — and an authorized distributor of test, measurement and network infrastructure solutions.", "email" : "sales@test4net.com", "foundingDate" : "2024", "image" : "https://test4net.com/hubfs/og-t4n.png", "logo" : "https://test4net.com/hubfs/test4net-logo-fixed-2.svg", "name" : "TEST4NET LLC", "sameAs" : [ "https://www.linkedin.com/company/test4net/", "https://t.me/TEST4NET", "https://www.facebook.com/people/Test4net-LLC/61591470856077/" ], "telephone" : "+374-60-88-88-99", "url" : "https://test4net.com/" } ``` ```json { "@context" : "https://schema.org", "@id" : "https://test4net.com/#website", "@type" : "WebSite", "inLanguage" : "en", "name" : "TEST4NET", "potentialAction" : { "@type" : "SearchAction", "query-input" : "required name=search_term_string", "target" : { "@type" : "EntryPoint", "urlTemplate" : "https://test4net.com/hs-search-results?term={search_term_string}" } }, "publisher" : { "@id" : "https://test4net.com/#organization" }, "url" : "https://test4net.com/" } ``` ```json { "@context" : "https://schema.org", "@type" : "BreadcrumbList", "itemListElement" : [ { "@type" : "ListItem", "item" : "https://test4net.com/", "name" : "TEST4NET", "position" : 1 }, { "@type" : "ListItem", "item" : "https://test4net.com/blog", "name" : "Blog", "position" : 2 }, { "@type" : "ListItem", "item" : "https://test4net.com/blog/lossless-ethernet-ai-fabrics-pfc-llr-cbfc", "name" : "Lossless Ethernet for AI Fabrics: PFC, LLR or CBFC — Which to Choose and How to Verify It", "position" : 3 } ] } ```