Cisco Centro de Datos de proxima generación, Cisco Data Center Nex Generation

Editor's Notes

1. With the rise of public cloud infrastructure providers, IT is faced with a new challenge – that of answering to the “public rate card”. It is only a matter of time before the executive committee, or even the board, is asking IT how they stack up against the economics of resources available through these public clouds. <click>Virtualization has already begun to dramatically remove the underlying complexity of IT infrastructure and operations. As a result, customers that are more aggressively adopting virtualization within their datacenters are achieving previously unattainable economics. <click> With broader virtualization, and the approach of resource pooling, automation of the ongoing operations of IT becomes a reality. It’s through the combination of utilization with virtualization, then applying automation, that the entire cost curve is being re-drawn. In fact, it is this same approach that is being applied by public cloud providers to attain their aggressive rate card. Internal IT can get there.
2. Unified Computing Relationship with Data Center 3.0 Consolidation, Virtualization, Automation: A Systems PerspectiveWithin the Datacenter there are Four planes that exist, the Process Automation, Network, Compute, and Virtualization layers. These planes contain heterogeneous components and choices.Unified computing is an architecture that transverses the planes and pre-integrates them into a solution that optimizes the intersection points.We believe that the unique value that Cisco offers comes at the intersection of these platforms. The intersection of the three bubbles brings clarity and focus to data center virtualization. At the Compute / Virtualization Intersection Three key areas to consider within virtualization: Memory, IO, and CPU. Depending on application performance any one of these will become the limiting factor to VM/Server ratios Memory – The Unified Computing solution has optimized memory – expanding the overall available memory to up to 384 gb per node IO optimization – increased IO including 10gb Unified Fabric – converging Ethernet and Fiber Channel traffic over a single cable CPU – utilizes the latest Intel chipset to expand performance The architecture optimizes all three areas allowing for more VM’s on a physical compute device At the Virtualization / Network Intersection Optimization of V-Motion calls have been integrated with the network. Management of the network directly to the virtual machine – removing previous limitations where network administrators could only manage to the physical device port Improved auditing and traceabilityAt the Compute / Network Intersection: Wire-once low latency Unified Fabric – 10gb per link Ethernet and Fiber Channel Significantly reduced components, removing points of failure All three platforms are enabled with Process Automation / data enablementThe management within the Unified Computing systems has fully open API’s (XML and industry standard) to integrate with Business Service Management solutions Provisioning workflow Event Monitoring CMDB data feedsPre-configured integration exists with key partners to speed time to deployment
3. Cloud Automation Pack at this stage is mainly delivered through a services engagement