Announcing MCP Toolbox Java SDK v1.0: Agentic data access for the enterprise
cloud.google.com·6d ago
Cloud & Infrastructure
Google Cloud Blog·14d ago·5 min read
Storage-optimized Z4D machine family, now GA, is designed for IO-intensive workloads
TL;DR
Google Cloud has launched the Z4D machine family, optimized for IO-intensive workloads, offering up to 84,000 GiB of local SSD storage and improved performance by up to 40% over previous models.
✦ Why It Matters
Consider migrating to Z4D instances to enhance performance for your IO-intensive applications today.
Key Takeaways
- Explore Z4D VMs for SQL and NoSQL workloads.
- Utilize bare-metal instances for low-latency applications.
- Leverage high local SSD capacity for AI/ML tasks.
- Migrate from Z3 to Z4D for better cost efficiency.
Full Summary
The new Z4D machine series in Google Compute Engine is designed for demanding workloads such as SQL, NoSQL, and AI/ML applications. It features both Virtual Machine (VM) and bare-metal instances powered by 5th Gen AMD EPYC processors.
Z4D provides significant enhancements, including up to 84,000 GiB of local SSD storage and performance improvements of up to 40% compared to the previous Z3 instances. The VMs come in two types, optimized for different workloads, and support high local SSD capacity per virtual CPU (vCPU).
Additionally, Z4D bare-metal instances allow for low-latency access to hardware, making them suitable for specialized applications. Customers have reported substantial performance gains, with some experiencing up to 70% reductions in processing time and improved throughput for data-intensive tasks.