SDxCentral Magazine - Issue 2
7-Eleven networks in Taiwan, Verizon’s CTO, OpenAI's networking, and AMD's head of networking tech
7-Eleven networks in Taiwan, Verizon’s CTO, OpenAI's networking, and AMD's head of networking tech
Standard update targets AI network bottlenecks
OCP provides a glimpse at the stack set to power Muse agent workloads
More power out and less power consumption
Switch pairs FBOSS OS with 102T chip for 100K+ GPU clusters
OCP introduces OSPA
‘Industry first’ portfolio looks to bridge copper-to-optical divide to drive AI infra scaling
Upscale’s Token Fabric to create network bridge for the dream of heterogeneous computing
Platform was designed with input from a phalanx of industry heavyweights
SoftBank's wide-area workload shift tech aims to sate AI compute demand regardless of power bottlenecks
Some 40 self-driving packages to launch by next summer
Network ASIC developer Zhanwang joins optics vendor and PC giant as general members
Carmen Palacios-Berraquero warns against setting post-quantum network standards too soon
Baseline framework looks to light up AI infrastructure adoption
And open interopability is key
Chinese lab taps Microsoft’s TileLang in push for simpler AI programming
EMEA network chief touts Ethernet lead; lauds OpenAI protocol as networking standard for scale-up and out
Nationwide deployment set to take place after initial fireworks festival trial
The recent trial hinted at what's to come
Understand the network behind Taiwan’s own backbone
Equinix and Fastly partnerships leverage APIs and edge-based compute for quick installations and uninterrupted football streams
Cisco engineer DIYs an MRC emulator
Chip giant embraces open chiplet interconnect standards to unlock heterogeneous compute
MetaRoCE reimagines Ethernet data transfers to fix network latency at AI scale
NTT, NEC, and 1Finity-backed effort secured international standard support
Nine years from debut, Kubeflow hits 260 million downloads
Allot-led coalition looks to PQC and QKD to combat future quantum threats
CS-4 promises 30-times faster AI workloads compared to GPU platforms with latency as low as two milliseconds
Lightweight Serial Interconnect effort looks to create open options to smash the AI memory wall