Closed Xuanwo closed 1 week ago
I also invite all @apache/opendal-committers to join the review.
Comparing add-vision
(271b697) with main
(de198cd)
⚡ 8
improvements
❌ 4
regressions
✅ 61
untouched benchmarks
:warning: Please fix the performance issues or acknowledge them on CodSpeed.
Benchmark | main |
add-vision |
Change | |
---|---|---|---|---|
❌ | buffer 256 KiB * 32 chunk |
183.6 ns | 212.8 ns | -13.71% |
❌ | buffer 256 KiB * 4 chunk |
154.4 ns | 183.6 ns | -15.89% |
❌ | buffer 4.00 MiB * 32 chunk |
154.4 ns | 183.6 ns | -15.89% |
❌ | buffer 4.00 MiB * 4 chunk |
154.4 ns | 183.6 ns | -15.89% |
⚡ | 256 KiB * 1000k chunk |
183.6 ns | 154.4 ns | +18.88% |
⚡ | 256 KiB * 1000k truncate |
245.8 ns | 216.7 ns | +13.46% |
⚡ | 256 KiB * 100k chunk |
183.6 ns | 154.4 ns | +18.88% |
⚡ | 256 KiB * 100k truncate |
245.8 ns | 216.7 ns | +13.46% |
⚡ | 256 KiB * 10k chunk |
212.8 ns | 183.6 ns | +15.89% |
⚡ | 256 KiB * 10k truncate |
245.8 ns | 216.7 ns | +13.46% |
⚡ | 256 KiB * 1k chunk |
183.6 ns | 154.4 ns | +18.88% |
⚡ | 256 KiB * 1k truncate |
245.8 ns | 216.7 ns | +13.46% |
Thank you all for the review, especially @tisonkun for the detailed comments and @xxchan for optimizing the expression. I'm going to merge this now.
Let's build "One Layer" for "All Storage"!
Which issue does this PR close?
Result of https://github.com/apache/opendal/discussions/5301
Rationale for this change
Establish VISION for "One Layer, All Storage"
What changes are included in this PR?
Update docs that refer to vision.
Are there any user-facing changes?
Not in the short term, but it will affect the entire future.