Managing virtualization / Leveraging application topologies
When visiting customers, we often speak with virtualization managers excited by the ROI their virtualization strategies are generating. The benefits include higher utilisation rates (often jumping from 10-15% to 80-85% for x86 machines), improved consolidation of infrastructure, and lots of cost savings in power and cooling.
But they also often express some common pitfalls: users are creating lots of new unsecured, unlicensed, unmonitored virtual machines across their environments without always following best practice, or even approved internal processes. Other blogs have started pointing to a general lack of management tools for virtualization.
Well, so what is Tideway going to do about it?
In short, it can be difficult to contain ‘virtualization sprawl’. And even harder to determine the possible negative impacts of over-virtualization on operational integrity.
Tideway likes to break the solution to this problem into two stages: First generate a detailed ‘currency audit’ of your entire virtual infrastructure. By this we mean a detailed configuration baseline of all the virtualization components in your environment. Second, map this virtualized infrastructure to the business applications they support.
Baselining the virtual environment
Usually the first stumbling block is “you first have to know what you have” in order point your discovery tools in the right direction. The agentless scanning technology of Tideway Foundation 6.3 avoids this need by generating a full-estate audit of all discoverable virtualized components, both for x86 and Unix-based platforms (one of our key strengths is the ability to give you this ‘rolled-up view’ across both platforms). The list of currently supported products is growing all the time.
Here is an example of the types of configuration information Foundation 6.3 can bring back for an ESX server with a virtual machine:

Note some of the key pieces of information the tool has reported on: an ESX server running one virtual machine, four virtual interfaces and their configuration, and one virtual switch and its configuration, including its IP address. From an operatational integrity perspective, we can spot immediately if we have any worrisome speed or duplex mismatches. All of this information is available through a single user interface for all of your virtualization products.
Mapping the virtual environment
The next phase is to contextualize these virtual components within the business applications they support. Here, Tideway’s advanced application modelling tools provide unique capabilities that no other virtualization management tool can offer: the ability to quickly determine the relationship between virtual hosts, the software products they contain, and the business applications they support.
To see what we mean, take a look at this application map from Foundation 6.3 for a bond trading platform (and several interdependent stock trading applications):

Here we see 4 inter-dependent business applications at the top, several supporting software packages (MS SQL, Apache, Tomcat, etc.) underneath in the light green , and then some hosts and their network interfaces all down to a single switch.
Two of these hosts are virtual. The physical containers for each are represented by the shaded boxes. If we drill into either the physical or virtual hosts, we can obtain detailed configuration information listed above. Not only do we know these virtual hosts are critical to this bond trading application, but we also know that they may impact the related stock trading applications.
Quick summary
The benefits of Tideway to the virtualisation manager can be summarized as:
- Control the sprawl through regular currency audits
- Map virtualized components to business applications
- Enforce best practice and standardization
The benefits to the business include:
- Reduced complexity
- Improved operational integrity and fewer outages
- Reduced manual processes and associated costs
To recap, customers generate the most value from Tideway in two phases. First by baselining their virtualization inventory with accurate count and configuration information, including conformance to standard builds, etc. In the second phase, they improve the transparency of their virtualized infrastructure by mapping components to business application topologies, and then spotting integrity or dependency issues that may cause problems. This combined functionality allows different virtualization technologies to be identified, tracked, and managed within an overall ITIL-based service management strategy.
