Skip to content

Commit a9c3ffa

Browse files
whelanBoydAdamGS
andauthored
Add Jan update (#29)
* Add Jan update Signed-off-by: Whelan Boyd <whelanboyd@gmail.com> * Clean up GPU support Signed-off-by: Whelan Boyd <whelanboyd@gmail.com> * Proposed changes Jan 2026 (#31) Signed-off-by: Adam Gutglick <adam@spiraldb.com> * Add latest GPU work Signed-off-by: Whelan Boyd <whelanboyd@gmail.com> --------- Signed-off-by: Whelan Boyd <whelanboyd@gmail.com> Signed-off-by: Adam Gutglick <adam@spiraldb.com> Co-authored-by: Adam Gutglick <adamgsal@gmail.com>
1 parent cc4bf9e commit a9c3ffa

File tree

1 file changed

+73
-0
lines changed

1 file changed

+73
-0
lines changed

src/content/blog/december-2025.mdx

Lines changed: 73 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,73 @@
1+
---
2+
title: "December & January Bulletin"
3+
date: "2026-01-30"
4+
authors: ["Community Team"]
5+
excerpt: "Overview of all work happening in Vortex"
6+
published: true
7+
---
8+
9+
The work mentioned in the previous bulletin to revamp Array evaluation to be fully lazy was [released](https://github.com/vortex-data/vortex/releases/tag/0.57.0) on January 6th. This happens by converting their execution to an Operator model that evaluates into Vectors (fully decompressed, zero-copy to Arrow representation). As a reminder, this work enables many more optimizations, and also provides unified abstractions for evaluating on
10+
different processor types (CPUs & GPUs).
11+
12+
## GPU Support
13+
14+
Speaking of, focus is now on support for reading Vortex files into GPUs. To achieve this, the team is
15+
16+
- Adding GPU decompression for existing kernels, as well as some new encodings optimized for GPU; and
17+
- Integrating with NVIDIA's CUDA toolkit for high performance I/O.
18+
19+
The first supported output types will be Arrow Device Arrays and cuDF, and a few key PRs have already landed.
20+
21+
- Add CUDA kernels for decoding bitpacked arrays on GPU ([#6145](https://github.com/vortex-data/vortex/pull/6145))
22+
- GPU filter kernel for primitive types ([#6188](https://github.com/vortex-data/vortex/pull/6188))
23+
- GPU filter kernel for decimals, strings and binary ([#6196](https://github.com/vortex-data/vortex/pull/6196))
24+
- GPU Scans ([#6199](https://github.com/vortex-data/vortex/pull/6199))
25+
26+
As with all of Vortex, these capabilities are fully exposed to plugins so advanced users can extend and customize for their own use.
27+
28+
## Around the Ecosystem
29+
30+
**DuckDB Labs** published [performance benchmarks](https://duckdb.org/2026/01/23/duckdb-vortex-extension) comparing Vortex (using the `vortex` extension) versus Parquet. Spoiler alert: Vortex reduced total query time by ~18-35%!
31+
32+
**Spice AI** began a blog series about building their data accelerato Cayenne with Vortex and Datafusion:
33+
34+
- [🌪️ Vortex: The Bet on Encoding-Efficient Columnar Storage for Hot Data](https://www.linkedin.com/posts/lukekim_datafusion-spiceai-data-activity-7417019189477126144-1TRe/)
35+
- [🔬 The Research Behind Modern Data Compression & Vortex](https://www.linkedin.com/posts/lukekim_datafusion-developers-ai-activity-7417649503291498496-TD5_/)
36+
- [🤖 Three Data Problems Vortex Solves for Applications and Agents in 2026](https://www.linkedin.com/posts/lukekim_vortex-efficient-columnar-storage-for-hot-activity-7419472524750798848-lZC6/)
37+
38+
**DataFusion:** Over the past two months we've released support for DataFusion 51, and then merged support for DataFusion 52.
39+
40+
- Expression conversion is now extendable, allowing users to push down custom UDFs or any other DataFusion expressions.
41+
- Support for schema evolution between files within the same table has improved dramatically, fixing many bugs and and allowing deeper pushdown over nested structs columns.
42+
- We now support tables that have more complex arrow types (Like Dict, REE), while also doing less work exporting from Vortex to Arrow.
43+
- Delegate some caching to to DataFusion's built in caching, allowing users to tune it using familiar configurations.
44+
45+
## Acknowledgments
46+
47+
We want to thank to anyone who has tried Vortex, provided feedback, asked question and filed issues.
48+
49+
The following contributed to the December & January releases.
50+
51+
```text
52+
Joe Isaacs
53+
Adam Gutglick
54+
Connor Tsui
55+
Nicholas Gates
56+
Alexander Droste
57+
Robert Kruszewski
58+
Alfonso Subiotto Marqués
59+
Andrew Duffy
60+
Cancai Cai
61+
Onur Satici
62+
Dmitrii Blaginin
63+
Dan King
64+
Baris Palaska
65+
godnight10061
66+
sherlockbeard
67+
Frederic Branczyk
68+
paultiq
69+
Pratham Agarwal
70+
Hao Huaijin
71+
Dave Bunten
72+
Harry Scholes
73+
```

0 commit comments

Comments
 (0)