We’re excited to share some exciting updates regarding performance improvements in Tulip. Over the past few weeks, our team has been diligently working behind the scenes to improve load times across the entire product. You’ll already see some of these improvements in r260, and we’re excited to announce that even more enhancements are on the way in r261 (LTS11).
The impact of these improvements extends far and wide, but it’s most notable in larger instances where load times used to be a bit sluggish. However, rest assured that these changes will benefit all users.
Here’s a quick rundown of what to look forward to:
We’ve optimized trigger editor load time, making them significantly snappier. Historically, we loaded trigger contents (like available tables) only when the trigger was loaded, causing reload delays. We’ve shifted this loading process to run in the background when the editor loads, resulting in an average 89% reduction in trigger load time.
We’ve also fine-tuned the load performance of “machine and device” app triggers. These triggers now load information about available machines and devices more efficiently. In our benchmarking, we achieved an astounding 99.2% reduction in load time for these triggers.
Many aspects of Tulip are displayed in lists (such as connectors, tables, and machines), and historically, these pages suffered from slower load times as product usage scaled up. We’ve made these pages a top priority, both during our recent hackathon and in ongoing projects. We did benchmarking at different load levels, and the difference is quite compelling -
- Users Page (2200 Users): 87% reduction
- Connectors Page (300 connectors): 80% reduction
- Machines Page (120 machines): 51% reduction
- Stations Page (2400 stations): 53.4% reduction.
- We also introduced an optional instance-level setting that reduces less critical information, bringing the load time down by 82.6%.
- Apps Page: With 1200 apps, we reduced load time by 68.1%.
These performance improvements translate into thousands of hours saved in load times over a year of usage. We believe these enhancements will greatly enhance your daily experience with our product.
As always, these improvements are a result of valuable feedback from our community. Please keep that feedback coming—we truly appreciate it!