Crucial to the success of a service, to IT, and to the business
Back in earlier times, Capacity Management was far easier. While many of the technologies we use today (including virtualization and shared resources) started decades ago on the mainframe, all the relevant work was done on the mainframe. The data needed for capacity management, to enable measurement and management of performance and capacity was easily available, well understood, and quickly put to use.
Things are different today. Mainframes and dumb terminals have given way to distributed, client-server environments. Those environments have grown in complexity as technologies that allow for clustering and virtualization bring us back closer to the mainframe model of decades before.
The missing piece of the puzzle for years in capacity management has been the quality of data that is available for the mainframe and the transactional view that is easy to gather and analyze from the rich transaction data that is available for the mainframe.
So, given that no one data set from the resources that make up today's data center contains everything that a capacity manager needs in order to make informed capacity management decisions, a different approach is required.
Angle over distance:
A key approach used in many fields is “angle over distance.” In other words, it’s more important to have the right approach (or the right angle) over getting as close as possible to the item (or for the capacity manager, the components) that you are analyzing. It’s easy to look at an individual server, for instance. Good performance and capacity data is available from most operating systems, databases, and server-based applications. What happens, though, when a transaction spans the Internet along with dozens of servers (web servers, middleware servers, application servers, indexing servers, etc.) and potentially a group of databases or a big data warehouse? It’s possible that service level agreements are not being met even though each component piece appears to working as designed. This is where “angle over distance” and 360° Capacity Management is crucial to the success of a service, to IT, and to the business.
From one angle:
It’s still crucial that all the hardware components be monitored and managed as one under-configured (or mis-managed) component could cause considerable performance and capacity problems. athene® is an important tool in the kit as it can capture performance and capacity data from all the hardware resources that make up today’s complex infrastructure.
(& the services that run on them)
Data can be captured and incorporated into a Capacity Management Information System (CMIS) by using athene Acquires or using agentless technology. Vital data such as business statistics, centralized storage and network data can be stored in the CMIS by using Integrator Capture Packs and provide many different angles for attacking capacity and performance challenges.
Once the data is stored in the CMIS, athene can be used to report, analyze, trend, and model the data in order to answer key capacity management questions. athene allows the capacity manager to plan for changes to the business (such as anticipated growth or merger/acquisition activity) or the services and applications that run on these hardware components.
From another angle:
Once the components are well-managed, it’s time to pay close attention to the transactions. Most service level agreements are written to ensure that a certain percentage of transactions complete within prescribed limits. In order to police those agreements, it’s necessary to capture key metrics from every production transaction. Sharepath is just the tool to do this, as it measures the end-to-end experience for every transaction and can alert the analyst on transactions that are not meeting (or are in danger of not meeting) those requirements.
(& the transactions that comprise them)
With Sharepath it’s possible to do much more than just measure, report, and alarm on end-to-end response times – it’s also capable of measuring the residence time for each hop in the path of the transaction. If transactions are performing poorly, it’s possible to see exactly where it’s doing so – this is a crucial piece of information and a necessary input to the Incident Management and Problem Management processes.
Capacity Management today has the same goals and desired outcomes as it did decades ago – providing adequate capacity at controlled costs, both now and in the future. The methods we use today and the tools needed for meeting these goals have changed due to the changing nature of IT services and the complexity of the IT infrastructure needed to run these services. Looking at performance and capacity challenges from all possible angles, 360°Capacity Management, gives us the ability to properly meet service level agreements and meet capacity management’s mission of providing adequate capacity at controlled costs, both now and in the future.