VMworld 2014: Software-Defined Data Center and vCloud Suite Roadmap

This session brought by Fausta Ibarra and Wayne Greene, both from VMware, was all about the roadmap and missions for SDDC and the vCloud Suite. I will share the highlights of this session with you in case you missed it.

The biggest part of the session was about technical barriers and limitations that will be moved with upcoming releases. These changes were divided in three categories: Compute, Networking and Storage and Availability. I’ll go into each of them briefly and share the most important bits with you.

Compute

vSphere 5.8 will support more than:

320 logical processors per host
4TB physical memory per host
4096 virtual processors
32 nodes per cluster
62TB per VMDK

Next, the following features were shared with the audience:

Fault Tolerance will finally start supporting Multi-Processor VMs (up to 4 virtual CPUs)
vMotions across vCenter Servers will be made possible
Long distance vMotion will be made possible (up to 100ms RTT)
vCloud Director functionality will be built in vSphere (ability to create virtual data centers and assign resources to it)

Networking

It wasn’t really clear what the roadmap items were for the networking part of this session, but I added the photos of the slides for your below anyway. In my opinion NSX will improve it’s integration with different VMware products in the future, and functionality will be added to the different NSX components like load balancers and firewalls. I can’t wait until a fully virtualized network will be implemented during my career. Public IP’s on your vCNS/NSX firewall and only physical spine and leaf switches.

Storage and Availability

This part of the session was all about the way storage should be approached. There will be hot-edge storage and cold-core storage. Hot being cpu and memory bound, having lower latencies and dominated by flash. Cold being capacity-centric, scale-out, multi-geo and extendable to the cloud.

Extra: Automation, Monitoring and Cloud

Not mentioned as a category, but a nice extra during the session.
Virtual Volumes will be the next-gen method of VM placement and things like large scale vMotions, application centric protection of workloads will be something VMware will focus on.

Some highlights from the roadmap items for vRealize Automation Center (the rest can be found in the photos below. Same goes for the rest of the products):

Multi-tenant Fabric Management
Policies for placement, network, security, scale and more
Lifecycle actions (versioning, updates, scale-out, backup)
Model release pipeline & push updates for continuous integration
Desktop services (Horizon View and Desktone)

Same approach for vCenter Orchestrator:

Phase out Design Center by simplifying vCAC lifecycle call-outs to external systems via VCO
Further integrate VCO with vCloud Suite (vRealize Suite?)
Remediate failures detected by vCOPS
Better error handling

The last subject covered Cloud Operations Management, IT Business Management and vRealize Air. Integration between different components and visibility of usage and costs will be further improved.

SaaS Management Center (SMC) that will be a single public portal used to manage and access all services and trials. Integrated support with VMware Support and Subscriptions
Migration solutions for easy moving between on-premise and public cloud
Service-Oriented