AWS
Reddit take
AI summary of the Reddit excerpts below — not a quote
AWS is a dominant cloud infrastructure provider that elicits polarized reactions from technical professionals. While it is praised for its high security standards and for eliminating the burden of physical hardware management, it is frequently criticized for its high costs, complex billing, and a support system described by some as 'support theater.' Opinions are mixed regarding its value proposition, with some users finding it cost-effective compared to hiring staff, while others view it as an expensive 'overkill' for many use cases.
Pros
- High infrastructure reliability compared to on-premise setups
- Superior security and MFA capabilities that are difficult to replicate locally
- Eliminates the need for managing physical hardware and data center rentals
- Effective for cold storage, removing risks associated with physical drive failure
Cons
- Cloud billing can be significantly more expensive than on-premise servers for certain workloads
- Customer support is often reported as unhelpful or slow during critical account issues
- Major outages can impact a significant portion of global internet traffic
- Account 'timeout' or suspension for billing issues can occur silently without prominent dashboard warnings
Caveats
- Billing warnings may be hidden in 'recommended actions' rather than displayed as urgent banners
- Internal Amazon teams reportedly face the same inflexible pricing as external customers
- One report highlighted a total data loss event despite having redundancy and backups configured
Best for
- Enterprises wanting to offload hardware maintenance and scaling
- Long-term data archival and cold storage
- Organizations requiring high-level security compliance and robust MFA
Avoid if
- You are running large relational databases where the cost of cloud services exceeds hardware and staffing
- You require immediate, high-quality human support for account-level crises
- Your project is small enough that AWS features would be considered 'overkill'
This summary uses balanced time weighting with about a 6-month half-life. The evidence is older (median age of ~402 days), meaning some specific service improvements or billing interface changes may not be reflected in the most recent comments.
Sentiment
20 positive (49%) · 7 mixed (17%) · 14 negative (34%)
Trust + time weighted score: +20% · raw score 15%
What redditors said (10 of 41)
AWS is overkill and it'll cost an arm and a leg
AWS for instance, you can get free resources to learn about the services and you can understand more about how they are configured. Get a free AWS account.
spin up a free tier AWS azure gcp etc account and build the stuff you audit on paper, make IAM roles, break your own access, turn on CloudTrail and read the logs.
Built-in integrations with cloud infrastructure (such as Azure or AWS) and SSO providers (such as Okta or Google Workspace) to automatically discover your internal systems, hardware, and third-party vendors.
Go over the Learn path and look up a video to get up to speed with terminology (granted that's way more sensible than what AWS does so this isn't that big of a deal anyway)
cant out printscreen from some referential AWS report .. i mean they rewrote the definitions of controls which is in second Column of part 4
I’ve noticed that AWS has a lot of free hands-on labs and sandbox environments available, but I haven’t found many equivalent options for Azure.
Client data is stored in AWS databases, managed by an external party who is ISO27001 certified → clearly in scope
MFA evidence screenshot or export showing MFA enforced on AWS, GitHub, and your IdP. Buyers check this first.
All our operations are running on cloud platforms (AWS and Azure).