Unblurring the Lines Between High Availability ,Disaster Recovery and Scalability
What are the biggest driver for enterprise to move to cloud ?
Cloud providers provides enterprise a manageable , secure, scalable, high-performance , efficient , elastic , High available , Fault tolerant and recoverable architecture in cost efficient manner .
Too many key words in one line right ? But that’s what cloud providers provider like AWS, Azure and GCP etc. been successful in enabling architectures which lives up to the expectation as per the aforesaid keywords.
TO achieve the value by migrating to cloud, it is very important to understand the difference between 3 Keyword HA, DR and Scalability
Although the terms scalable, HA(Highly available) , DR(Disaster Recovery) are not new they all were relevant in traditional on-prem architecture as well but often I have seen these terms are used interchangeably and hence create confusion which leads to Implementation delays and surprises at the time of unpredicted load or any failure .
SO. In the this blog series
- Let’s understand what is the objective of Scalable, HA and DR
2. Challenges with Traditional HA, DR and scalability Approach
3. How Scalability , HA and DR ia achiived in cloud its merits
Objective of Scalable, HA and DR architecture
Larger objective of all these 3 is same i.e consistently good customer experience in case of any unpredicted event.
Unpredicted event broadly can be categorized of 3 Types
· Failure at Infra level — Power, network , Server, software failure
· Natural Calamity — Earthquake , Flood in the datacenter location
· Unpredicted load: Infra is not ready to cope with the sudden increase in the demand
So Enterprise Architect job is to make sure business has zero to minimum impact in case of above unavoidable events
Lets Understand now Specific objective of Scalable, HA and DR
Highly Available Architecture make sure that an application continues to operate when one or more underlying component fails . SO in HA it is important to configure a failover system that can handle the workload if primary system is failed . So HA is achieved by adding redundant component
I have often seen people are confused that HA architecture will take care of unpredicted load . HA strategy is not designed make sure that your system will be available for any amount of Load . HA is responsible for making sure there is minimum downtime in case of any failure
DR Strategy is required for Business continuity in case any catastrophic disaster .Although DR strategy is to make sure Availability in case of catastrophic disaster but Its is much more than HA strategy . DR strategy includes
· Use of Alternate geographical location not just the redundancy at the infra or data center level
· DR also focuses instantizing the service not just failover
· DR strategy involves more poeople and process to. Manage
Scalable Architecture is one that can handle increasing numbers of workloads without affecting the performance (response time and throughput). So Scalable architecture strategy is to make sure your system responds as quickly as possible for increasing loads with no impact on end customer experience .
I have Created Below Table to summarize
In next blog I will explain what are the Challenges with Traditional HA, DR and scalability Approach and How Scalability , HA and DR ia achiived in cloud its merits
