F5 Synthesis: High-Performance Services Fabric

#SDN #Devops #Cloud #SDAS #SDDC To achieve the economy of scale necessary to ensure no application is deprived of critical application services, you need to abstract resources.

In the early days of cloud computing we talked a lot about how the economy of scale offered by cloud was achieved mainly through abstraction of resources. Compute, network and storage resources were abstracted and pooled together such that they could be provisioned as services on-demand.

That economy of scale ensured that the cost of using those services decreased, making them affordable for even the smallest of organizations.

In the data center, however, similar economy of scale has been difficult to achieve because the abstraction at the network layers has remained elusive. SDN has recently emerged as a front-runner in the data center as a provider of that abstraction, turning disparate network elements into a cohesive fabric of network resources, dynamically adjusted to deliver the best performance and availability to any application that might be delivered over its formerly rigid pipes.

But SDN doesn't provide an easy answer for layer 4-7 (application) services. Application services must be able to match the economy of scale offered by server virtualization and SDN / NFV if economy of scale in the manner of cloud computing is to be realized.

 

That's why F5 Synthesis contains as a key component a high-performance services fabric. And I do mean high performance. Its aggregate connection capacity is 9.2B - more than one connection for every person on the planet or 3 times the capacity needed to connect every Internet user across the globe*. It can support up to 80 virtual instances per device and has an aggregate throughput of 20TB. All-active clustering thanks to ScaleN with a focus on application services ensures fault and resource isolation, so neighboring tenants can't starve your services.  The services fabric provides application- and service-level control and failover to deliver industry-leading reliability. Workloads can be moved across the fabric without interrupting other services and can be scaled to meet the business demand.

F5's high-performance services fabric supports traditional and emerging underlay networks. It can deployed a top traditional IP and VLAN-based networks, works with SDN overlay networks using NVGRE or VXLAN (as well as a variety of less well-known overlay protocols) and integrates with SDN network fabrics such as those from Cisco / Insieme, Arista and BigSwitch among others.

The services fabric model enables consolidation of services onto a common platform that can be deployed on hardware, software or in the cloud, reducing operational overhead by standardizing management as well as deployment processes to support continuous delivery efforts. By sharing service resources and leveraging fine-grained multi-tenancy, the cost of individual services is dramatically reduced, enabling all applications regardless of size to take advantage of services that are beneficial to their security, reliability and performance. 

Additional Resources:

1. F5 Synthesis: The Time is Right 
2. F5 and Cisco: Application-Centric from Top to Bottom and End to End 
3. F5 Synthesis: Software Defined Application Services 
4. F5 Synthesis: Integration and Interoperability

[*] Based on Sept 2013 Internet user population of 2.4B http://www.internetworldstats.com/stats.htm