[#WhatILearntSeries] Infrastructure Utilization, Capacity Planning and Forecasting
High availability, low latency and security are the most needed factors for a quality service when it comes to any customer-facing applications.
If companies lose track of what is happening or what might happen in their infra environment , performance problems and capacity shortage may result in the loss of revenue, reduced productivity, and degraded customer experience. A proper frame work needs to be followed ensuring cost is being optimized without compromising the needed quality. An optimized infrastructure with good monitoring and tracking measures in place protects organizations from unexpected bills, and the total costs will drop accordingly.
Based on my understanding so far, whole capacity management process can be broken into 3 sub-categories:
👉 Business level : Understand the future demands of users/customers
👉 Service level : Make sure services can be offered at needed SLA
👉 Resource level : Make sure supporting resources can meet the SLA promised
But how do we achieve it?
One must create an infrastructure with enough capacity to handle existing and future demands replying on answers to below key questions:
🔹 Are applications performing according to users’ expectations?
🔹 If I add xx volume of new client, can it handle the expected traffic?
🔹 Is there enough storage available?
🔹 What information is available for Latency, Message Size, Throughput, Work being done by transaction, concurrent users
🔹 What will my system needs to look like in xx months with xx percent of growth?
🔹 How should my system perform at peak hours and at all possible peak times of the year?
.. and list goes on until #gapanalysis is complete!
Above all, critical component is monitoring the success once the plan is in place. Note that there is really no real end date to this process and you will constantly adjust the infrastructure for size, speed, reliability etc unique to your use cases.
