Xsight Labs has joined the Open Secure AI Alliance, part of The Linux Foundation.
As #AI workloads scale exponentially, securing data in motion across high-speed data center networks is just as vital as raw performance and power efficiency.
By collaborating with industry leaders across hardware, software, and #cloudinfrastructure, we are helping drive open, standardized security standards for the next-gen AI compute environments.
Together, we’re building the fast, efficient, and secure data paths powering the AI era: secureaialliance.org#SecureAIAlliance#AIInfrastructure#Semiconductors#Networking#AISecurity@linuxfoundation
ALT Xsight Labs Linux Foundation project Open Secure AI Alliance
The countdown to #OCP2026 is on!
Xsight Labs is hitting the floor at the OCP Global Summit in San Jose, Oct 12-15.
We’re ready to showcase how we're powering the next era of AI data centers, and we might just be dropping some big news while we're there 👀
Drop by our booth or reach out to get an invite to our suite to see what’s next for high-performance open #networking!
👉 Details & Registration: opencompute.org/summit/globa…#Semiconductors#DataCenter#AIInfrastructure#OpenHardware
Yesterday, we showed you the Add-in card and today we are #unboxing the E1 Server.
Neil O'Rourke takes you through the unboxing, rear/front I/O connectivity, internal motherboard architecture, and #PCIe expansion features of the Xsight E1 #DPU Server:
🔹 Front I/O & Management: 2x 400G QSFP112 ports, RunBMC, 1GbE/USB OOB, 5x UART ports
🔹 Internal Specs: 4x DDR5 DIMMs, M.2 SSD Linux boot, optimized oversized heatsink
🔹 PCIe Bays: Dual FL/DH front expansion bays with dedicated aux power routing
Watch the full video to explore how the E1 Server simplifies deployment for high-density #datacenter environments: piped.video/watch?v=LEIxHuYm…#Networking#HardwareEngineering#ServerArchitecture
Let's unbox the Xsight E1 #PCIe Add-in Card!
Neil O'Rourke breaks down the key hardware features and installation steps for the E1 800G DPU Add-in Card.
🔌 Front I/O
Dual QSFP112 400G, Micro-SMA precision timing, OOB Ethernet
🚀 Host Interface
PCIe Gen 5 x16 + MCIO expansion
🛠️ Server integration
Server compatibility & aux power setup across major platform vendors
Watch the full unboxing here: piped.video/WJC3PiTB-0M#DataCenter#Networking#Unboxing#PCIeGen5
#AI should save you money or make you money - period.
The catch? Calculating the true Total Cost of Ownership (#TCO) means juggling complex variables across power, cooling, server offload, and lifecycle management.
We made it easy.
In her latest blog post, @TheRealJKaplan breaks down DPU economics and reveals the methodology behind our E1 TCO Calculator showing exactly how the E1 DPU delivers measurable savings to your data center architecture.
Quantify your #infrastructure savings: xsightlabs.com/blog/how-we-t…#DPUEconomics#CloudArchitecture#DataCenters#TechInnovation
Our E1 #DPU Takes the Spotlight on Arm's Neoverse CSS N4 Page: arm.com/products/cloud-datac…
As next-generation #AI and hashtag#cloud data centers demand massive leaps in infrastructure capabilities, it's exciting to see the Xsight #E1 800G DPU prominently featured on @Arm's official #Neoverse CSS N4 product page as a prime example of Arm's compute ecosystem in action.
Some things never go out of style, like building open, efficient, and fully programmable infrastructure for #AI and beyond.
Here is what we consider Kinda Chic!
ALT ▶️ Next-Level Reach: Sending your silicon straight into space on a SpaceX rocket.
ALT
▶️ Industry Firsts: Launching the very first 800G DPU on the market.
ALT Open Infrastructure: Building open, efficient, programmable infrastructure for AI and beyond.
ALT Extreme Efficiency: Running a fully programmable 12.8 Tbps switch at under 200 Watts.
Where in Europe is Werner Schueler? Starting today, he is hitting the road for the ultimate Xsight European Tour!
If you are attending the top-tier tech events across Germany this September, keep an eye out for @wschueler to discuss the future of #AIinfrastructure, #cloud networking, and next-gen #silicon:
🇩🇪 Digital X | September 8, 2026 - Cologne
🇩🇪 Deutsche Telekom Satellite Day 2026 | September 10, 2026 - Bonn
🇩🇪 Cloud X Summit (by STACKIT) | September 17, 2026 - Wiesbaden
🇩🇪 Bits & Pretzels (The Founders Festival) | September 28–30, 2026 - Munich
🇩🇪 IONOS Summit | September 30, 2026 - Berlin
#XsightOnTour#DigitalX#BitsAndPretzels#CloudXSummit#IONOSSummit#CloudInfrastructure#Networking#AISilicon@bitsandpretzels
Werner Schueler is thrilled to join key industry leaders and visionaries, including keynote speaker Dr. Angela Merkel, to explore the future of innovation, security, and AI transformation.
At Xsight Labs, we’re constantly pushing the boundaries of data center #infrastructure to accelerate AI workloads and deliver ultra-scalable processing power.
Heading to #DigitalX in Cologne? Let’s connect and talk about how next-gen connectivity is shaping the future of tech!
📍 Where: #Rheinauhafen, Cologne
📅 When: September 8, 2026
Also, it's at a chocolate museum...
@wschueler
The 48 hours that rewrote Israel's startup rankings.
Wonderful, Upwind and HiBob all revealed sharply higher price tags this week, and one rising star may already be on its way out the door.
calcalistech.com/ctechnews/a…
Power constraints and rigid data planes are stalling next-gen data centers. We teamed up with Netberg to solve both with the new Aurora X2 Switch Series.
By pairing Xsight’s X2 programmable #silicon with Netberg’s open hardware, operators get software-defined flexibility without sacrificing power efficiency.
⚡ 12.8 Tbps @ <200W - Max performance, low thermal overhead
🧠 Line-Rate Telemetry - Fine-tune packet brokering and load balancing
📊 64MB Shared Buffer - Microburst protection for high-density AI clusters
Whether you're building high-density 400G spine networks or scalable 100G access layers, open #programmablesilicon is driving the next era of infrastructure.
👇 Dive into the architecture behind the Aurora X2 series: xsightlabs.com/blog/netberg-…#HardwareEngineering#CloudComputing#NetworkArchitecture#AIInfrastructure
🎉 We are happy to announce the first members of the inaugural P4 Ambassador cohort! p4.org/2026-ambassador-cohor…
We received many applications but these nine individuals embody what we’re looking for ⬇️
Most #AIFactories do not have enough #DPUs relative to their #GPU count, and as a result, GPUs are severely underutilized. Despite billions spent on $30,000+ accelerators, real-world cluster telemetry shows average GPU utilization sitting far below potential, often idling 15% to 50% of the time during training runs and dropping even lower in unoptimized enterprise environments.
Enter the E1: The first 800G DPU on the market.
We launched our interactive TCO Calculator so you can see the math behind offloading #infrastructure tasks to the #E1.
Plug in your node count, power rates, and GPU costs to model your 4-year ROI, CAPEX/OPEX savings, and recovered GPU hours or swipe through for Hyperscaler vs. Enterprise reference runs.📊 Model your cluster: xsightlabs.com/tco-calculato…#AITraining#DataCenter#ComputeEfficiency
What’s hotter than the #silicon at Hot Chips Symposium? ☕️ The coffee powering our team at Booth A2 on the last day. 💪 We’re bringing our latest ultra-low power, high-throughput interconnect tech to the floor.
Stop by to chat next-gen cloud #infrastructure, see our hardware, or just debate whether standard heat sinks look cool (Spoiler: ours do 😉)
@hotchipsorg
✅ Badge scanned
✅ Notebook full of architecture notes
✅ Mind blown by next-gen #bandwidth
✅ Stopping by Booth A2 to see Xsight Labs in action
Don't leave Hot Chips Symposium with FOMO.
Swing by our booth to meet the team, catch our latest tech in action, and score some prime swag before it’s gone.
@hotchipsorg