NetIX Extends Peering Reach Across Eight Russian MSK-IX Hubs
📅 - NetIX has expanded direct connectivity into Russia by linking its platform to MSK-IX across eight major metropolitan hubs, from Moscow to Vladivostok. The deployment gives NetIX members broader access to Russian networks and hosted content, while allowing MSK-IX participants to reach NetIX locally through nearby points of presence rather than relying on more centralized interconnection paths today.
The expansion covers Moscow, St. Petersburg, Rostov-on-Don, Samara, Kazan, Ekaterinburg, Novosibirsk and Vladivostok. For carriers, content providers, hosting companies and other networks using NetIX, the useful part is geographic reach. Traffic destined for Russian networks can be exchanged [...][... Check source for end of article ...]
📅 - Lenovo Claims NVMe Memory Tiering Can Cut VMware Costs by 45% - Lenovo says VMware Cloud Foundation Advanced Memory Tiering on ThinkAgile VX can cut memory costs by up to 45% by using NVMe as a secondary tier alongside DRAM. The approach is aimed at private cloud operators trying to raise VM density and run larger workloads without expanding expensive memory footprints at the same rate across existing infrastructure.
The number is attention-grabbing, but it needs qualification. Lenovo's estimate is based on its own cost analysis of selected 1TB memory configurations using DRAM and NVMe against systems configured entirely with DRAM. Actual savings depend on workload and configuration. So this is not a blanket 45% reduction in VMware infrastructure [...]
📅 - HPE Data Center Networking Revenue Jumps 112% on AI Demand - HPE's data center networking revenue more than doubled in its fiscal third quarter, rising 112.2% year over year to $382 million as demand for AI infrastructure broadened beyond servers. Routing revenue surged 270% to $788 million, while server sales climbed 35.3% to $6.8 billion, underscoring how accelerated computing is reshaping infrastructure spending across enterprise infrastructure.
The networking numbers are the more revealing part of the quarter for data center operators. Hewlett Packard Enterprise reported total networking revenue of $2.9 billion, up 74.9%, with campus and branch revenue increasing 31% to $1.4 billion and security revenue rising 75.6% to $281 million. Data center [...]
📅 - GTZHost Publishes Technical Tutorial on Fine-Tuning Meta's Llama 3 on NVIDIA A100 GPU Servers - As enterprise adoption of open-source Artificial Intelligence continues to accelerate, customizing base models for specialized business tasks has become a primary objective for data science teams. Meta’s Llama 3 stands out as one of the most capable open-source Large Language Models (LLMs) available today. However, because it functions as a general-purpose model out of the box, enterprises must fine-tune it on proprietary datasets to handle domain-specific workflows such as specialized medical reporting, proprietary code generation, or automated financial analysis.The Computational Demands of LLM Fine-Tuning Fine-tuning a modern LLM requires massive parallel c
📅 -