How Cloud Resiliency organized in Microsoft Azure


Disaster avoidance and Disaster recovery are two concepts used in many industrial areas. The two terms look alike but they are somewhat different based on the situation that we use them in. Sometimes disaster avoidance is also called resiliency. In information technology, these two terms are mainly used in system engineering and system designing areas. So, what is meant by terms resiliency and recoverability? Resiliency expresses the ability to operate services even in a disruptive system event. Recovery means actions needed to perform to make a system work again when a disruption causes a system to fail. For example, resiliency is like building structures to stay stable in an earthquake situation, on the other hand, recoverability means if some disaster happens to the structure how we are going to correct that damage. So, now as we know the meaning of the two terms, let’s move to our main topic, How Cloud Resiliency is organized in Microsoft Azure. Cloud consists of a larger number of data centers. These data centers are similar to on-prem data centers but huge in scale. Cloud is mainly built using virtualization technologies same as on-prem data centers. Therefore, the resilience options that the cloud has similar to what we use in on-prem data centers. However, different cloud providers use different names to market their products. In this article, let’s discuss a few key resilience concepts offered by the Azure cloud service.

First let’s have some idea on a few key terms, starting with regions, region is a logical name given to a group of data centers that are placed somewhat close to each other. Regions are organized under geographies. Geography is a boundary defined based on political, regulatory, and location factors. Under regions there are multiple zones, a zone has at least one data center, a zone has its own dedicated power supply and network service. Zones contain multiple fault domains and update domains. Fault domain identifies a group of hardware that have common cooling, power, and network service. We can think of a fault domain as a server rack. Update domain identifies a group of hardware that gets fabric level update at the same time. Azure ensures that they do not update two or more update domains at the same time. These policies apply as same to zone level. In this paragraph we identify a few commonly used keywords that we need when talking resiliency options in Azure, let’s discuss a few resiliency options from next para. 

Let’s start with availability sets, it is a logical group that spans through few fault-domains and update-domains. The recommended setting is to include 2-3 fault domains and 5-20 update domains within an availability-set. After setting up, then we can place over service, under the availability set. Users do not have the ability to select which fault or update domains to place their service. Azure cloud handles that part therefore Azure highly recommends creating separate availability sets for different tasks. Availability set guarantees 99.5% availability.  

Next, let’s talk about Availability Zone. This concept is all about placing your services and resources in multiple zones. This ensures that in the event of zonal level disruption still your service can operate as normal. Availability zones guarantee 99.9% service availability. Under current configurations, a service can be deployed to one of three availability zones. Selection of the availability zone should be carried out by customers. Therefore, when architecting your applications, you have to be really careful about where to place services. Regions that support availability zones contain at least three zones. 

Next, we have paired regions, if you wish to have regional level resiliency then this service is for you. Regional level Azure service could fail due to mainly natural disasters, war situations but this type of situation occurs rarely and normally you have enough time to react before that scale of an event happens.

These are the commonly used resiliency services. However, there are many more resiliency options we can use with Azure cloud. Hope to discuss those options in future posts.

References

Comments

  1. Dulanga Kithmini29 November 2020 at 18:01

    Nice write-up. As per my understanding, in cloud computing, applications are involved with many types of resources and the applications are depending on internal and external services. There can be issues due to the failures in those services or defective software. So, resiliency should be involved with how these failures can be detected and recovered. Has Azure come up with a mechanism to address this context too?

    ReplyDelete
    Replies
    1. Osura Weerasinghe6 December 2020 at 17:02

      Yes Dulanga, there are various offerings in azure to address different situations as you mentioned. For example, we can use Azure application insights to identify behaviors of application that are running in the cloud, Azure Monitor is another service we could use to monitor infrastructure and application running on Azure, we have a security center to monitor security-related activities. Azure also offers best practices and recommendations with the Azure advisor.

      Delete
  2. Santhoopa Jayawardhana6 December 2020 at 11:37

    As a leading cloud service provider it is good that Azure has addressed the issues related to availability ensuring high availability of it's services.

    ReplyDelete
    Replies
    1. Osura Weerasinghe6 December 2020 at 16:49

      Yes, it is must to do, because of the high competitiveness in cloud business all the cloud providers need to come up with their own availability strategies to stay in the business.

      Delete
  3. Rajitha Gamage10 December 2020 at 12:42

    It's very informative. keep it up!

    ReplyDelete

Post a Comment

Popular posts from this blog

Network Traffic Identification and Classification with Machine Learning

Image
A computer network should consist of four primary objectives: Fault tolerance, Scalability, Security, and Quality of Service (QoS). These objectives make computer networks efficient, reliable, and suitable to satisfy users' requirements. Network administrators try to continuously improve these properties to give a better experience for their customers. There are various practices network admins can implement to achieve these four main objectives related to computer networking. Knowing about the network is one of the initial and basic activities that we can commonly see in those practices. Network monitoring is the process we use to get to know about the network. Network traffic monitoring plays a significant part in computer networking. It has been used from network resources management to network security-related activities.  Network monitoring is an umbrella term that is used to combine a few other processes/techniques. Network traffic identification and classification (TIC) is a...
Read more

MalLocker.B

Image
New  Malware Affecting A ndroid Users. Malware has been there for a very long time from the beginning of computer systems. Any computer system could be a target for a malware attack. The common intention of any malware is to cause damage to computer systems. Before the era of smart devices, malware was mainly designed to attack servers, personal computers, and likewise. However, with the rapid increase of smart devices, malware attack surfaces grew rapidly. Smart devices became a comparably easy target for threat actors because of the extensive user base, less knowledge of users about the device, users trying to add various applications without checking their background, along with many other reasons. These reasons enable attackers to quickly and easily spread malware to smart devices. Ransomware is one type of malware that causes huge losses for both business and personal computer systems in current times. This type of malware holds the computer system in a captive state until t...
Read more