<!channel> I'm excited to announce the public rele...
# important-announcementsj
Jason Bosco
12/08/2021, 4:37 AM<!channel> I'm excited to announce the public release of v0.22 of Typesense! 🎉 🎉
Some key highlights:
✅ 20-30% reduction in memory usage
✅ 20-25% improvement in search performance
✅ 5x speed-up in update performance
✅ Configurable rules to auto-apply filters. Eg: "Best-rated restaurants" can automatically trigger the filter "rating > 4.5"
✅ Configurable typo tolerance based on word length
✅ Ability to index special characters verbatim or split search tokens based on them
✅ Server-side caching for heavy queries
✅ Improved relevance for multi-field searches
And a host of other UX improvements and bug fixes.
A big thank you to everyone who contributed ideas, tested out the RC builds and helped bring this release to life! 🙏
More detailed information here: https://typesense.org/docs/0.22.0/api/#what-s-new
Oh also, I just noticed we've crossed 400 members in our Slack community! 🎉 🎉
❤️ 20
🙌 1
👍 26
🎉 35
Jason Bosco
12/08/2021, 4:38 AMIf you're on Typesense Cloud, please email support@typesense.org with your cluster ID to upgrade to this version.
If you're self-hosting (especially a multi-node HA cluster), please read through these important upgrade instructions that's specific for this particular version upgrade: https://typesense.org/docs/0.22.0/api/#upgrading
m
Masahiro Aoki
12/08/2021, 5:35 AMCongrats !!! 🎉 🎉 🎉
🙏 1
c
Carl
12/08/2021, 7:35 AMGreat work guys 💪
🙏 1
b
bnfd
12/08/2021, 8:43 AMGreat!! Thank you Kishore and Jason!
🙏 1
c
Chaitanya Sharma
12/08/2021, 4:49 PMwould love to readup about these performance wins, is that written about where ? @Kishore Nallan
k
Kishore Nallan
12/09/2021, 5:21 AMThere was no single magic bullet. We reduced wasteful allocations, adopted more cache-friendly internal data structures and reducing lock contentions. A series of
benchmark -> identify hot spots -> improve perf👍 1
j
Janaka Abeywardhana
12/31/2021, 9:06 AM👏🏾