The rise and fall of Silicon Valley Bank: a tragic tale of greed and high-interest rates

Silicon Valley Bank (SVB) has gone the way of the dodo, comparable to the collapse of the Great Recession. SVB lost a whopping $1.8 billion in a securities sale and tried to raise more capital to regain its footing, but it only fueled doubts about its financial stability.

The resulting bank run was classic drama: anxious clients scrambling to get their deposits back immediately. Then, prominent venture capitalists urged companies to withdraw from SVB, and shares plummeted by about 60% on Thursday, and another 65% on Friday, culminating in trading being stopped. As if it was scripted, SVB's fate was sealed with remarkable speed, and now it's defunct.

SVB primarily catered to tech startups, and as the sector experienced a downturn, customers withdrew their funds, resulting in a cash shortage. To make matters worse, 93% of SVB's $161 billion in deposits were uninsured. In conclusion, SVB's collapse will have far-reaching consequences. Let's keep abreast of how the situation will unfold, shall we?

Meta's P92: The Future of Social Media or Just Another Flop?

Meta, the largest player in the social media game, is now building a standalone decentralized social network for text-based updates. The move comes as Twitter's decline has paved the way for other platforms to build next-generation replacements.

The app codenamed P92, is still in its earliest stages, and there are no details about its release. Legal and regulatory teams are already investigating potential privacy concerns around the app. Building a decentralized network could give Meta the opportunity for its new app to interoperate with other social products. This would enable users to choose from a variety of ranking algorithms that better reflect their desired experience.

However, there is no profitable, global-scale decentralized network yet. Basic social-network functions like following users become complicated when accounts are located across a vast network of servers.

"Trillions" of Messages? Discord Says, "No Sweat!"

You won't believe how a simple database migration solved the problem of storing trillions of messages for Discord. In 2017, they shared their journey of migrating from MongoDB to Cassandra, but as their database grew to 177 nodes and trillions of messages, Cassandra became a high-maintenance system with frequent issues that caused latency problems. One reason for this was the uneven distribution of messages in the partitioning, which created hot partitions and slowed down the cluster. Maintenance tasks also caused trouble, like falling behind on compactions and JVM garbage collection.

After experimenting with ScyllaDB, a Cassandra-compatible database written in C++, the team decided to migrate all of their databases. Though it took time to prepare and ensure a smooth transition, the migration to ScyllaDB led to a garbage collector-free environment and improved performance. Bo Ingram goes into greater detail in his article about the challenges and benefits of switching to ScyllaDB!

How did you like today’s email?