Datacenter in the clouds

What Do We Do to Reduce Complexity in the Next Generation of the Datacenter?

Over the next 24 months, the use of private/hybrid cloud environments to run SQL Server is projected to grow 82% (see chart below). These cloud deployments will surpass non-cloud, bare-metal environments 42% vs. 37%. It's pretty safe to say that this transition will have a dramatic impact on the operational processes of most IT organizations. It's also pretty clear that most IT organizations will be desperate to avoid the pratfalls that accompanied rapid adoption of server virtualization infrastructure.

See also: Need for Control Drives SQL Server Platform Choices

Sure, that infrastructure-centric focus to application management lowered physical server capital expenses. But many organizations saw those savings wiped out by skyrocketing software and operational costs because of the need for higher-edition software versions and more labor-intensive and complex processes.

SQL Server Database Environments
Note: Multiple responses allowed | Data: InformationWeek 2014 State of Database Technology Survey of 956 business technology professionals, January 2014

So how will we reduce the complexity of managing these on-premise and private/hybrid cloud ("hybrid datacenter") SQL Server environments? It's going to take a movement away from traditional infrastructure-focus management tools. This new class of application-centric, distributed application management solutions will leverage new lightweight application virtualization technologies (e.g., containers) to decouple an application instance from its underlying host and IT infrastructures. The new focus will be on application instance mobility/portability, automation, and orchestration.

Recent research by Forrester provides insight into the need for this new focus:

  • Application workloads won't be virtualized to save on hardware, but to make them portable. The easiest way to manage SQL Server workloads across on-premise and private/hybrid cloud environments is to make them portable. That is, you need the ability to move a workload from one host (e.g., cloud VM) to another host (e.g., on-premise bare metal). The use cases for this mobility include managing performance-related issues and achieving better availability and/or business continuity.
  • Application workloads need to dictate the best virtualization platform, not the other way around. This is an acknowledgement that all applications are not created equal and that the early virtualization method of shoe-horning every application into a virtual machine was not productive. Automating the placement of workloads according to quality of service (QoS) metrics optimizes infrastructure resources and reduces costs.
  • Application workloads must be service-level resilient and disaster-tolerant. The ability to orchestrate SQL Server workload business continuity now becomes paramount in a hybrid datacenter model. At any given time, a workload could be running on a different host, but service level agreements (SLAs) must be met.

For SQL Server users, the benefits of this new class of distributed application management solutions will be:

  • Increased agility through workload portability. This lets the application workload dictate the best infrastructure platform. This workload portability lets you take full advantage of your existing infrastructure or create a cloud infrastructure utility consisting of heterogeneous physical and virtual servers optimized to meet specific application workloads needs.
  • HA/DR for every application instance. Because of enhanced workload portability, every application instance can by failed-over from host to host or site to site, enabling service-level delivery automation with full dependency coordination.
  • Guaranteed SLAs for performance & availability. You'll be able to intelligently tie an application's workload QoS requirement to business SLA requirement. Every instance can be protected with its own independent, portable QoS policy to ensure that availability and performance SLAs are always achieved. Workloads will be pro-actively monitored and automated IT processes for all application instances will automatically load-balance workloads across infrastructure to ensure performance even when in a degraded state.
  • Simplified lifecycle management. Because application instances are now decoupled from the underlying infrastructure, you'll be able to embrace change faster with less impact, resources, and cost. You'll also see reductions in maintenance and patch management, which creates more opportunities for innovation.
  • Lowest software & infrastructure costs. You'll be able to quickly find the optimal SQL Server software cost/compliance inflection point and drive infrastructure and licensing utilization rates higher.

Reducing complexity through application mobility/portability, automation, and orchestration will speed workload portability, enhance service continuity, and lower operational costs. And that's what we're all looking for in the generation of the datacenter.

TAGS: Hybrid Cloud
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.