DeepSeek looks to offload simple LLM tasks to save billions of parameters
'Conditional memory' concept moves simple lookups from expensive GPU memory to cheaper CPU RAM
'Conditional memory' concept moves simple lookups from expensive GPU memory to cheaper CPU RAM
Seeking proposals to scale up circuits for transceivers, accelerators, and other chip types
Extended maturities mark Broadcom’s confidence in VMware subscriptions and AI-driven revenues
Researchers propose low-latency topologies and processing-in-network as memory and interconnect bottlenecks threaten inference economic viability
Assessment collaboration tackles shadow AI and more
Interconnection Academy's partnership with Nomad Futurist to deliver industry-certified courses to professionals and students
Latest moves continue HPE's Juniper Networks integration efforts
2026 projected to be an SSD dry year, especially with the boon of key value caching
Next-gen Rubin platform will see DPUs help control all network I/O to and from the compute node
Observe to keep us safe from agents
First memory chips. Then lasers. Now, copper looks to be the next commodity to pose ‘systemic risk’ to the digital infrastructure space
CES showcase sees Lenovo offer pay-as-you-go access to AI inference infrastructure
Osmos uses agentic AI to process raw data into analytics that can be used in various data lakes
The deal boosts SoftBank's AI plans
Datadog has its day, and Dell on the up
This includes migrating 'key internal workloads into the hyperscaler
Part of firm's strategy to build AI-native security platforms aligned with enterprise needs that are evolving alongside cybersecurity threats
The goal is to improve operational efficiency and equipment uptime
Offering verifies network and accelerator integrity and can spin up validated clusters “in minutes'
AWS doubles down on custom silicon and the looming memory crisis to come
Japanese giant joins effort building an open, interoperable Ethernet architecture for next-gen workloads
Korean giant also working with SanDisk
The cybersecurity firm strengthens protection at the AI interaction layer as it sees prompts as 'the new malware'
NCCL support comes to GCP, alongside launch of A2UI open source project