⛅️ Accelerated capacity expansion to meet client needs. The new server base on HPE Synergy 12000 delivers up to 1.5–2.5 times more performance and allows for rapid resource scaling without "waiting in line" during load spikes.

⛅️ Made network operation more stable under high loads. The transition to Cisco Nexus 9000 with 100 Gbps within the cloud removes internal traffic bottlenecks (DB clusters, ETL, microservices) and reduces network incidents due to a unified stack. I wrote about issues related to internal network limitations here.

⛅️ Data storage became even more reliable. The introduction of HPE Primera A670 and expansion of backup circuits provide all-flash for critical databases and applications, plus additional backup capacity to avoid hitting limits and maintain deeper recovery history.

⛅️ Made AI infrastructure more flexible. A GPU pool (RTX 6000 Ada and RTX 5090) with hourly billing allows model training, rendering, and inference without capital investment in hardware — pay only for actual usage.

⛅️ Future plans. By mid-2026, the company will complete network unification and put HPE Primera into production — current clients will automatically gain access to a more powerful and stable platform.

Source: habr.com/ru/companies/cloud4y/articles/1001442/

#Cloud4Y #cloud #infrastructure #virtualization #GPU #backups #ITforbusiness