The hottest IT Infrastructure Substack posts right now

Beyond Earth Technologies is looking for innovative projects that can help with living on other planets and also have real opportunities for business now.
If you're a scientist or inventor working on things like energy, robots, or habitats, you can apply to join their program by January 31st.
You can share this opportunity with friends who have great ideas, and it only takes a few minutes to apply.

CrowdStrike made a big mistake by rolling out an untested update to all users at once, causing millions of computers to crash. They need to treat configuration updates like real code and test them properly.
Delta Airlines faced huge losses because it didn’t have backup systems in place when the CrowdStrike update went wrong. Having spare systems or a better contingency plan could have minimized disruptions.
Microsoft should improve its recovery methods after crashes, possibly by adopting an automatic system recovery strategy. Learning from other platforms could help avoid these issues in the future.

Migrating authorization logic from an old system to a new one can take a long time and requires careful planning to avoid errors.
Each part of a business can manage its own authorization rules, making it easier for them to control access based on their specific needs.
As systems grow, it's important to keep improving and adapting to new challenges, like optimizing runtime decisions and better analyzing access logs.

Denial of service (DoS) attacks aim to overwhelm a system with traffic, rendering it inaccessible. Robust security operations center capabilities are crucial for detecting and mitigating these attacks effectively.
Microsoft Sentinel offers tools like analytics rules, incident management, and threat intelligence integration for detecting and responding to DoS attacks in real-time.
To mitigate DoS attacks, organizations can leverage network traffic monitoring, DDoS protection integration, and incident response playbooks offered by Microsoft Sentinel.

Data migration is the process of moving information from one place to another, like relocating files when changing devices. It involves transferring various types of data, such as documents and databases, to ensure everything is in the right spot.
Migrations can be complex and risky, often causing errors or service disruptions if not done carefully. This makes it crucial for companies to have good planning and oversight to avoid losing important data or negatively affecting users.
There are many reasons to migrate data, such as upgrading technology or meeting new security regulations. Companies often need to adapt to growth or changes in the market, which can lead to costly and lengthy migration projects.

Get a weekly roundup of the best Substack posts, by hacker news affinity:

You can create XPath queries to use Data Collection Rules (DCRs) for the Azure Monitor Agent.
A shortcut trick to create XPath queries is using Event Viewer on a Windows system.
In Event Viewer, filter log files, enter Event IDs, check the XML tab for the XPath query, and use it in your DCR.

Providers take a percentage or tiered savings from your reserved instances savings.
Check if the mechanism of RI coverage is automated or manual to match your risk tolerance.
Consider how buyback works and if providers credit back lead time or offer instant sales.

There are two similar code paths for fetching posts by hashtag and user ID, causing duplication that can complicate the codebase. Simplifying these could make the code easier to manage.
Using a single index for both types of queries can reduce complexity and allow for easier future features, like querying by other criteria, without increasing code duplication.
Collapsing code paths not only streamlines current processes but also makes it easier to implement new features or queries in the future, reducing overall development effort.

Microsoft and OpenAI are launching a massive $115 billion supercomputer project called Stargate by 2028. This shows a huge investment in AI technology.
AWS plans to spend $150 billion on new data centers over the next 15 years to meet the rising demand for AI tools. This indicates that many companies are getting ready for a future filled with AI.
NVIDIA is making advancements in AI with its 'AI factories' and next-gen chips. They are pushing boundaries in technology and aim to help develop artificial general intelligence.

To set up Google Cloud Platform (GCP) for a company, you first need to create an organization. You can do this by signing up with either Google Workspace or Cloud Identity.
After creating the organization, the next step is to create users and groups. This is done through the admin console using your admin account.
Once users are set up, you can create projects and manage permissions. This allows different users to have specific access, like creating storage buckets, based on their roles.

Automated Moving Target Defense (AMTD) makes it hard for cyber attackers by constantly changing the rules.
AMTD hides vulnerabilities from attackers and reduces the time they spend inside your network.
Integrating AMTD with SASE solutions can reduce attack surface, limit attack time, and strengthen your cybersecurity.