Managed Services

What can we learn from recent AWS Outage

On November 28, 2020, the AWS outage took down a large portion of the internet and affected thousands of companies like Adobe, ConnectWise, etc., whose sites experienced outages in a domino effect. The outage was reported by many news sources like ZDnet and CNBC. The outage, which lasted several hours, occurred in AWS’ Northern Virginia, US-East-1 region, and the cause of the outage was later explained by Amazon in detail as having to do with a failure in an attempt to change capacity on Amazon’s famous S3 storage that resulted from an incorrect input to a simple command entered by a team member.
While server outages are nothing new, this latest AWS outage does offer everyone a few simple lessons to learn regarding vulnerability of mission critical systems.
Lessons to learn from recent AWS Outage:
  • Migration to AWS/Azure Cloud will not Automatically Make your System More Reliable: While there are important reasons to migrate to big boys’ cloud, reliability is not necessarily one of them. Reliability is the result of planning and proper execution, and does not entirely rest with your IaaS provider. A firm may be better served by a Managed Services Provider that understands its business and manages its servers whether on premise, in a data center, or in the cloud for reliability, resilience, and availability.
  • Proper Processes and Procedures Must be in Place to Avoid Incidences: As we learned, AWS outage, that lasted several hours and took down services of thousand of companies, was caused by just a wrong parameter passed to a simple command. Without proper checks and balances in place, mistakes like these do happen and often have far reaching consequences.
  • Migration to AWS/Azure Cloud will Not Automatically Lower Cost: In most cases, companies trade capital expenditures (fixed costs) for recurring monthly expenses (variable costs) when they migrate to the cloud. However, they must also include management of the infrastructure in their analysis over 3-5 year window before deciding which option is more cost effective.
ddadmin

Share
Published by
ddadmin

Recent Posts

Application Ring-Fencing and Storage Control

In our previous edition of this newsletter, we had discussed how ZTNA (Zero Trust Network Access) can…

3 years ago

Japan – Tokyo 2020 Olympics

Just as the games were kicking off, officials disclosed that the usernames and passwords of…

3 years ago

United States – Electronics Arts (EA)

Hackers have leaked an estimated 751GB of compressed EA data containing FIFA 21 source code…

3 years ago

United States – Republican National Committee (RNC)

Russian-aligned nation-state cybercriminals hacked into the Republican National Committee last week. Initially dismissive of the…

3 years ago

Zero-Trust Network Access

As users continue to work from anywhere and IoT devices flood networks and operational environments,…

3 years ago

United States – Mercedes Benz USA

Mercedes-Benz USA has disclosed a data breach impacting some of its US customers. The data…

3 years ago