The design decisions behind ScyllaDB’s elasticity and efficiency
With ScyllaDB X Cloud, several multiyear engineering projects converge to optimize elasticity and price performance – especially for dynamic and storage-bound workloads. Each represents a significant shift from the original ScyllaDB architecture. What’s the real story behind all of these changes? What other options did we consider, and what’s most interesting about the paths we took?
Join us as Tim Koopmans interviews ScyllaDB Co-Founder and CTO Avi Kivity on the motivations, design decisions, and next steps for projects such as:
- Tablets: What were the greatest challenges in dynamically distributing and rebalancing data with tablets
- Autoscaling: How do you ensure scaling events react in real time to changing workloads while avoiding over correction
- Compression: Why did we double down on storage bound workloads and what else is in store
- Storage: What combination of features lead to 90% storage utilization in ScyllaDB X Cloud
