Capacity and scalability testing and planning has become more important than ever with the increase of global deployments with significant web 2.0 and user performance requirements. Most importantly, your on-line revenue and profitability is also dependent on your web site’s availability.
The capacity planning challenge is consistently adding the right hardware at the right time to minimize capital cost without affecting availability. The scalability challenge is ensure that the architecture is in place so that the capacity added yields the appropriate benefit.
The answer to rock solid availability is to both establish and implement a capacity plan and test and correct scalability bottlenecks. Availability is not adding more servers. Instead it is translating your business requirements to the capacity and architecture will likely need to change to handle both growth and spikes in demand. Often we find poor architecture and lack of capacity planning leading to diminished availability and excessive capital investments. To combat this, we will help you change, refine, and develop your new or existing architecture while keeping your existing site running in a way that maximizes your capital investment and reduces your risk of downtime.
Deliver and implement capacity plans based on key parameters from monitoring the performance of your critical applications, your expected growth and spikes in demand.
Find bottlenecks before your customers do with automated scalability testing for web sites and the supporting databases.
Support you in updating or creating the best in class architecture for your mission critical applications.
Empower your team to use our capacity planning model and methodology.
Reduce risk to rollout and upgrade new and existing web sites and applications prior to production
Deliver on service level agreements, which include performance, availability, hardware and software costs by accurately predicting system capacity and scalability requirements.
Use load analyses to not only identify the point where system failure occurs, but also correct the problem.
We track detailed data, including download and transaction round-trip time, given various load conditions so that we can predict when you will have a problem well before you actually do.
Being pioneers and industry leaders in Capacity Management services for years and with expertise SAAS, Crescendant is working on development of “Captius” which helps customers to forecast demand, improve utilization of existing capacity, and to plan and purchase necessary and sufficient infrastructure resources.
Captius will be implemented using Software as a Service [SAAS] model in-order to provide better and timely service to customers.
When completed Captius will offer these service over web.
A lot of our customers have grave concerns about the capability of their applications, infrastructure and support teams to handle additional load due to increased traffic or new application features. Capacity assessment is intended to give a quick assessment of where company stand in those areas. Following data is captured for a period of a week or a month:
- From Infrastructure teams: infrastructure landscape diagrams, server metrics (CPU, memory, i/o, disk utilization, network utilization), access logs, application debug logs. From Application Support teams: organization charts, lists and access to information like call statistics, incident reports, service requests, planned and unplanned downtime reports.
- From Performance Management teams: Detailed performance and availability data from the set of monitoring nodes used for this application.
Captius will examine all this data thoroughly, applying rigorous numerical methods to come up with Capacity Assessment to give a composite rating from 1 to 5 to your application.
With this rating in hand, management can take short term action to rectify any gaps in the operations of the application.
Will continue with gathering data:
- From Infrastructure teams: as in Capacity Assessments
- From Application Support teams: as in Capacity Assessments
- From Development teams: application architecture diagrams, location of code repositories, lists of debugging hooks in applications, application use cases, major release to production plans for the year ahead
- From Business sponsors and management: major business events planned for the year ahead (e.g. the rollout of Windows Vista causing a spike in driver downloads, rollout of online banking in Brazil), the major SLA objectives for the application
Using these data Captius will gauge the maximum load that can be handled by the current infrastructure.
Captius will use business forecasts and rigorous numerical methods such as regression analysis of system metrics and log files to determine the expected growth in traffic, and hence, the expected load on the systems.
Captius will analyze support teams handling of incidents and service requests to predict what call volumes they can handle.
With all the collected data in hand, Captius issues ratings on a scale of 1 to 5, and comprehensive plans for each of these areas:
Infrastructure: Using the results from load testing and load forecasting, and knowing the prices, delivery times, and part numbers of your hardware, software and network vendors, we come out with a detailed quarter by quarter plan for procurement of hardware, software and network capacity that includes server model numbers, exact number of processors and memories, number of interfaces, etc.
Development: We provide recommendations for improving all the performance flaws exposed during load testing such as optimizing garbage collection, changing database queries for faster execution, etc. We also provide recommendations for where to add hooks in the code to improve troubleshooting of performance and availability issues.
Support: We suggest ways to automate support tasks, change server configurations, and optimize the operations of your support team based on our extensive operations experience. We suggest personnel changes or enhancements and additional training where required.
Business: By looking at prior business forecasts and actual server load and traffic, we can determine the accuracy of business forecasts. We will then create guidelines and questionnaires so that business sponsors begin feeding more accurate data to IT teams.
Data: How to automate the collection of data so that successive Capacity Planning cycles take less effort.
At the end of this process, company will be armed with a Capacity Plan telling where it is in terms of Capacity, and what changes and expenditures it needs to make in each of the above areas to have the best utilization and availability of capacity.
Capacity Assessment tells you where a company is and Capacity Planning tells you where company is going, and how to get there. Capacity Operations is a service that manages the process of getting from here to there.
Captius will take ownership for:
- Scripting and automating the execution of load tests and monitoring from various nodes.
- Automating data collection from all possible sources as identified in your Capacity Plan.
- Preparing reports for monthly management reviews to show the progress of execution of the Capacity Plan in each of the areas (Infrastructure, Development, Support, Business and Data) for the application.
- Driving your vendors and suppliers to make sure that infrastructure is procured, installed, and commissioned on time.
Captius reduces the tasks performed by existing Support and Infrastructure teams and reduces the operating costs.
About Crescendant | Contact | Employment | Privacy | Terms of Use | © Crescendant 2012 |