vortex-data
vortex
Blog
Docs
Changelog
Blog
Docs
Changelog
Overview
Branches
Benchmarks
Runs
Performance History
Latest Results
only a read lock for current_array_ref Signed-off-by: Daniel King <dan@spiraldb.com>
dk/buffer-pool
11 hours ago
only a read lock for current_array_ref (#6631) ## Summary SharedArray uses a Mutex as of #6453. This interacts particularly badly with `scalar_at` by way of `is_valid`. In particular, all calls to `scalar_at` on the given array are serialized. It seems reasonable to allow multiple threads to simultaneously read the validity. This PR permits that by using an RwLock. ## Testing I have a Vortex-based system which recently pulled in a version of Vortex that includes #6453. With that version, my profile looks like this: https://share.firefox.dev/4tO94Ki <img width="1920" height="1080" alt="Screenshot 2026-02-20 at 4 26 28 PM" src="https://github.com/user-attachments/assets/77f24c44-268e-4dd8-8bc7-ec2c193443b6" /> With this PR applied (i.e. with an RwLock in place of the Mutex), it looks like this: https://share.firefox.dev/4aNPZzt <img width="1920" height="1080" alt="Screenshot 2026-02-20 at 4 27 25 PM" src="https://github.com/user-attachments/assets/dcf2fde3-e35a-413b-a6e6-2b4053e05653" /> CPU utilization is high, and runtime is ~7s. By comparison, with a Mutex, CPU utilization is low and runtime is ~18s. Signed-off-by: Daniel King <dan@spiraldb.com>
develop
12 hours ago
GlobalData should not be passed as exclusive reference to thread-local functions Signed-off-by: Nicholas Gates <nick@nickgates.com>
ngates/duckdb-global-mut
12 hours ago
chore(deps): update dependency ray to v2.54.0 [security]
renovate/pypi-ray-vulnerability
13 hours ago
DuckDB lifetimes (#6620) Closes #6581 with different naming convention This PR removes the janky `owned: bool` flag and uses the same C API wrapping model as openssl. We macro `Foo` and `FooRef` structs for owned and borrowed pointers respectively. Owned pointers invoke their destructor on drop. --------- Signed-off-by: Nicholas Gates <nick@nickgates.com> Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk> Co-authored-by: Nicholas Gates <nick@nickgates.com>
develop
13 hours ago
Fix up Signed-off-by: Nicholas Gates <nick@nickgates.com>
ji/ngates/duckdb-lifetimes
13 hours ago
only a read lock for current_array_ref Signed-off-by: Daniel King <dan@spiraldb.com>
dk/rwlock-in-shared-array
13 hours ago
Add an `extension` module to `vortex` (#6624) ## Summary Tracking Issue: https://github.com/vortex-data/vortex/issues/6618 Since `vortex-dtype` and `vortex-scalar` have merged into `vortex-array`, and since planned extension logic now covers all 3 of dtype, scalar, and array, we need to pull out the existing extension dtype logic so that all 3 of these modules can access it. This PR adds a `vortex::extension` submodule to `vortex` that contains the `ExtVTable` (which was previously the `ExtDTypeVTable`), and a sub-submodule `dtype` that contains things like `ExtDTypeRef`. We should hope to add other sub-submodules `scalar` and `array` that would have `ExtScalarValueRef` and `ExtArrayRef`. I am wondering if we need to figure out the naming (please read the discussion in #6618) right now. Do we care that much right now or can we just change it later? ## API Changes Adds `vortex::extension` (which is pulled from `vortex_array::extension`), and moves some logic around extension dtypes and extension vtable into this new submodule. AFAIK nobody is actually using the new extension dtype so I think it should be fine to break? There is still a concern though if we release this version and someone decides to start implementing `ExtVTable`: we need to add a lot more methods to that ## Testing N/A since this is just moving code around. --------- Signed-off-by: Connor Tsui <connor.tsui20@gmail.com> Signed-off-by: Nicholas Gates <nick@nickgates.com> Co-authored-by: Nicholas Gates <nick@nickgates.com>
develop
16 hours ago
Active Branches
Dk/buffer pool
last run
11 hours ago
#6635
CodSpeed Performance Gauge
-65%
Convert &mut to & for DuckDB GlobalData
last run
12 hours ago
#6634
CodSpeed Performance Gauge
-27%
chore(deps): update dependency ray to v2.54.0 [security]
last run
13 hours ago
#6633
CodSpeed Performance Gauge
0%
© 2026 CodSpeed Technology
Home
Terms
Privacy
Docs