Skip to content

Commit 8b0d09d

Browse files
nvliyuansameerz
nvliyuan
and
sameerz
authored
[DOC] update the download page for 2504 release [skip ci] (#12497)
This PR updates download docs for 25.04.0 release. Signed-off-by: liyuan <yuali@nvidia.com> --------- Signed-off-by: liyuan <yuali@nvidia.com> Co-authored-by: Sameer Raheja <sameerz@users.noreply.github.com>
1 parent 0f29ae7 commit 8b0d09d

2 files changed

Lines changed: 116 additions & 23 deletions

File tree

‎docs/archive.md‎

Lines changed: 91 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -5,6 +5,97 @@ nav_order: 15
55
---
66
Below are archived releases for RAPIDS Accelerator for Apache Spark.
77

8+
## Release v25.02.1
9+
### Hardware Requirements:
10+
11+
The plugin is tested on the following architectures:
12+
13+
GPU Models: NVIDIA V100, T4, A10/A100, L4, H100 and B100 GPUs
14+
15+
### Software Requirements:
16+
17+
OS: Spark RAPIDS is compatible with any Linux distribution with glibc >= 2.28 (Please check ldd --version output). glibc 2.28 was released August 1, 2018.
18+
Tested on Ubuntu 20.04, Ubuntu 22.04, Rocky Linux 8 and Rocky Linux 9
19+
20+
NVIDIA Driver*: R470+
21+
22+
Runtime:
23+
Scala 2.12, 2.13
24+
Python, Java Virtual Machine (JVM) compatible with your spark-version.
25+
26+
* Check the Spark documentation for Python and Java version compatibility with your specific
27+
Spark version. For instance, visit `https://spark.apache.org/docs/3.4.1` for Spark 3.4.1.
28+
29+
Supported Spark versions:
30+
Apache Spark 3.2.0, 3.2.1, 3.2.2, 3.2.3, 3.2.4
31+
Apache Spark 3.3.0, 3.3.1, 3.3.2, 3.3.3, 3.3.4
32+
Apache Spark 3.4.0, 3.4.1, 3.4.2, 3.4.3, 3.4.4
33+
Apache Spark 3.5.0, 3.5.1, 3.5.2, 3.5.3, 3.5.4, 3.5.5
34+
35+
Supported Databricks runtime versions for Azure and AWS:
36+
Databricks 11.3 ML LTS (GPU, Scala 2.12, Spark 3.3.0)
37+
Databricks 12.2 ML LTS (GPU, Scala 2.12, Spark 3.3.2)
38+
Databricks 13.3 ML LTS (GPU, Scala 2.12, Spark 3.4.1)
39+
40+
Supported Dataproc versions (Debian/Ubuntu/Rocky):
41+
GCP Dataproc 2.1
42+
GCP Dataproc 2.2
43+
44+
Supported Dataproc Serverless versions:
45+
Spark runtime 1.1 LTS
46+
Spark runtime 2.0
47+
Spark runtime 2.1
48+
Spark runtime 2.2
49+
50+
*Some hardware may have a minimum driver version greater than R470. Check the GPU spec sheet
51+
for your hardware's minimum driver version.
52+
53+
*For Cloudera and EMR support, please refer to the
54+
[Distributions](https://docs.nvidia.com/spark-rapids/user-guide/latest/faq.html#which-distributions-are-supported) section of the FAQ.
55+
56+
### RAPIDS Accelerator's Support Policy for Apache Spark
57+
The RAPIDS Accelerator maintains support for Apache Spark versions available for download from [Apache Spark](https://spark.apache.org/downloads.html)
58+
59+
### Download RAPIDS Accelerator for Apache Spark v25.02.1
60+
61+
| Processor | Scala Version | Download Jar | Download Signature |
62+
|-----------|---------------|--------------|--------------------|
63+
| x86_64 | Scala 2.12 | [RAPIDS Accelerator v25.02.1](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.02.1/rapids-4-spark_2.12-25.02.1.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.02.1/rapids-4-spark_2.12-25.02.1.jar.asc) |
64+
| x86_64 | Scala 2.13 | [RAPIDS Accelerator v25.02.1](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.02.1/rapids-4-spark_2.13-25.02.1.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.02.1/rapids-4-spark_2.13-25.02.1.jar.asc) |
65+
| arm64 | Scala 2.12 | [RAPIDS Accelerator v25.02.1](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.02.1/rapids-4-spark_2.12-25.02.1-cuda11-arm64.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.02.1/rapids-4-spark_2.12-25.02.1-cuda11-arm64.jar.asc) |
66+
| arm64 | Scala 2.13 | [RAPIDS Accelerator v25.02.1](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.02.1/rapids-4-spark_2.13-25.02.1-cuda11-arm64.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.02.1/rapids-4-spark_2.13-25.02.1-cuda11-arm64.jar.asc) |
67+
68+
This package is built against CUDA 11.8. It is tested on V100, T4, A10, A100, L4 and H100 GPUs with
69+
CUDA 11.8 through CUDA 12.0.
70+
71+
### Verify signature
72+
* Download the [PUB_KEY](https://keys.openpgp.org/search?q=sw-spark@nvidia.com).
73+
* Import the public key: `gpg --import PUB_KEY`
74+
* Verify the signature for Scala 2.12 jar:
75+
`gpg --verify rapids-4-spark_2.12-25.02.1.jar.asc rapids-4-spark_2.12-25.02.1.jar`
76+
* Verify the signature for Scala 2.13 jar:
77+
`gpg --verify rapids-4-spark_2.13-25.02.1.jar.asc rapids-4-spark_2.13-25.02.1.jar`
78+
79+
The output of signature verify:
80+
81+
gpg: Good signature from "NVIDIA Spark (For the signature of spark-rapids release jars) <sw-spark@nvidia.com>"
82+
83+
### Release Notes
84+
* Support the Spark functions Bin and TruncDate
85+
* Support group-limit optimization for ROW_NUMBER
86+
* Improve Spark metrics: Print the batch size information to executor log
87+
* Refine filter push down to avoid double evaluation
88+
* Grab the GPU Semaphore when reading cached batch data with the GPU to avoid a GPU OOM case
89+
* Add an option to disable measuring buffer copy to improve large shuffle large partition serialization
90+
* For updates on RAPIDS Accelerator Tools, please visit [this link](https://github.com/NVIDIA/spark-rapids-tools/releases)
91+
* Upgraded statically linked CUDA toolkit to 12.8, which includes support for GB100 GPUs
92+
93+
Note: There is a known issue in the 25.02.1 release when decompressing gzip files on H100 GPUs.
94+
Please find more details in [issue-16661](https://github.com/rapidsai/cudf/issues/16661).
95+
96+
For a detailed list of changes, please refer to the
97+
[CHANGELOG](https://github.com/NVIDIA/spark-rapids/blob/main/CHANGELOG.md).
98+
899
## Release v25.02.0
9100
### Hardware Requirements:
10101

‎docs/download.md‎

Lines changed: 25 additions & 23 deletions
Original file line numberDiff line numberDiff line change
@@ -18,12 +18,12 @@ cuDF jar, that is either preinstalled in the Spark classpath on all nodes or sub
1818
that uses the RAPIDS Accelerator For Apache Spark. See the [getting-started
1919
guide](https://docs.nvidia.com/spark-rapids/user-guide/latest/getting-started/overview.html) for more details.
2020

21-
## Release v25.02.1
21+
## Release v25.04.0
2222
### Hardware Requirements:
2323

24-
The plugin is tested on the following architectures:
24+
The plugin is designed to work on NVIDIA Volta, Turing, Ampere, Ada Lovelace, Hopper and Blackwell generation datacenter GPUs. The plugin jar is tested on the following GPUs:
2525

26-
GPU Models: NVIDIA V100, T4, A10/A100, L4, H100 and B100 GPUs
26+
GPU Models: NVIDIA V100, T4, A10, A100, L4, H100 and B100 GPUs
2727

2828
### Software Requirements:
2929

@@ -49,13 +49,15 @@ The plugin is tested on the following architectures:
4949
Databricks 11.3 ML LTS (GPU, Scala 2.12, Spark 3.3.0)
5050
Databricks 12.2 ML LTS (GPU, Scala 2.12, Spark 3.3.2)
5151
Databricks 13.3 ML LTS (GPU, Scala 2.12, Spark 3.4.1)
52+
Databricks 14.3 ML LTS (GPU, Scala 2.12, Spark 3.5.0)
5253

5354
Supported Dataproc versions (Debian/Ubuntu/Rocky):
5455
GCP Dataproc 2.1
5556
GCP Dataproc 2.2
56-
57+
5758
Supported Dataproc Serverless versions:
5859
Spark runtime 1.1 LTS
60+
Spark runtime 1.2
5961
Spark runtime 2.0
6062
Spark runtime 2.1
6163
Spark runtime 2.2
@@ -69,41 +71,41 @@ for your hardware's minimum driver version.
6971
### RAPIDS Accelerator's Support Policy for Apache Spark
7072
The RAPIDS Accelerator maintains support for Apache Spark versions available for download from [Apache Spark](https://spark.apache.org/downloads.html)
7173

72-
### Download RAPIDS Accelerator for Apache Spark v25.02.1
74+
### Download RAPIDS Accelerator for Apache Spark v25.04.0
7375

7476
| Processor | Scala Version | Download Jar | Download Signature |
7577
|-----------|---------------|--------------|--------------------|
76-
| x86_64 | Scala 2.12 | [RAPIDS Accelerator v25.02.1](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.02.1/rapids-4-spark_2.12-25.02.1.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.02.1/rapids-4-spark_2.12-25.02.1.jar.asc) |
77-
| x86_64 | Scala 2.13 | [RAPIDS Accelerator v25.02.1](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.02.1/rapids-4-spark_2.13-25.02.1.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.02.1/rapids-4-spark_2.13-25.02.1.jar.asc) |
78-
| arm64 | Scala 2.12 | [RAPIDS Accelerator v25.02.1](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.02.1/rapids-4-spark_2.12-25.02.1-cuda11-arm64.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.02.1/rapids-4-spark_2.12-25.02.1-cuda11-arm64.jar.asc) |
79-
| arm64 | Scala 2.13 | [RAPIDS Accelerator v25.02.1](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.02.1/rapids-4-spark_2.13-25.02.1-cuda11-arm64.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.02.1/rapids-4-spark_2.13-25.02.1-cuda11-arm64.jar.asc) |
78+
| x86_64 | Scala 2.12 | [RAPIDS Accelerator v25.04.0](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.04.0/rapids-4-spark_2.12-25.04.0.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.04.0/rapids-4-spark_2.12-25.04.0.jar.asc) |
79+
| x86_64 | Scala 2.13 | [RAPIDS Accelerator v25.04.0](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.04.0/rapids-4-spark_2.13-25.04.0.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.04.0/rapids-4-spark_2.13-25.04.0.jar.asc) |
80+
| arm64 | Scala 2.12 | [RAPIDS Accelerator v25.04.0](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.04.0/rapids-4-spark_2.12-25.04.0-cuda11-arm64.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.12/25.04.0/rapids-4-spark_2.12-25.04.0-cuda11-arm64.jar.asc) |
81+
| arm64 | Scala 2.13 | [RAPIDS Accelerator v25.04.0](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.04.0/rapids-4-spark_2.13-25.04.0-cuda11-arm64.jar) | [Signature](https://repo1.maven.org/maven2/com/nvidia/rapids-4-spark_2.13/25.04.0/rapids-4-spark_2.13-25.04.0-cuda11-arm64.jar.asc) |
8082

81-
This package is built against CUDA 11.8. It is tested on V100, T4, A10, A100, L4 and H100 GPUs with
82-
CUDA 11.8 through CUDA 12.0.
83+
This package is built against CUDA 11.8. It is tested on V100, T4, A10, A100, L4, H100 and GB100 GPUs with
84+
CUDA 11.8 and CUDA 12.8.
8385

8486
### Verify signature
8587
* Download the [PUB_KEY](https://keys.openpgp.org/search?q=sw-spark@nvidia.com).
8688
* Import the public key: `gpg --import PUB_KEY`
8789
* Verify the signature for Scala 2.12 jar:
88-
`gpg --verify rapids-4-spark_2.12-25.02.1.jar.asc rapids-4-spark_2.12-25.02.1.jar`
90+
`gpg --verify rapids-4-spark_2.12-25.04.0.jar.asc rapids-4-spark_2.12-25.04.0.jar`
8991
* Verify the signature for Scala 2.13 jar:
90-
`gpg --verify rapids-4-spark_2.13-25.02.1.jar.asc rapids-4-spark_2.13-25.02.1.jar`
92+
`gpg --verify rapids-4-spark_2.13-25.04.0.jar.asc rapids-4-spark_2.13-25.04.0.jar`
9193

9294
The output of signature verify:
9395

9496
gpg: Good signature from "NVIDIA Spark (For the signature of spark-rapids release jars) <sw-spark@nvidia.com>"
9597

9698
### Release Notes
97-
* Support the Spark functions Bin and TruncDate
98-
* Support group-limit optimization for ROW_NUMBER
99-
* Improve Spark metrics: Print the batch size information to executor log
100-
* Refine filter push down to avoid double evaluation
101-
* Grab the GPU Semaphore when reading cached batch data with the GPU to avoid a GPU OOM case
102-
* Add an option to disable measuring buffer copy to improve large shuffle large partition serialization
103-
* For updates on RAPIDS Accelerator Tools, please visit [this link](https://github.com/NVIDIA/spark-rapids-tools/releases)
104-
* Upgraded statically linked CUDA toolkit to 12.8, which includes support for GB100 GPUs
105-
106-
Note: There is a known issue in the 25.02.1 release when decompressing gzip files on H100 GPUs.
99+
* Support approx_count_distinct
100+
* Support group by on binary type
101+
* Support ArrayPosition function
102+
* Support Databricks 14.3 ML LTS (without support for Deletion Vector reads in Delta Lake)
103+
* Support Slice
104+
* Enable Hive text writer
105+
* Refine split-retry logs when out of memory happens to expose the real reason
106+
* Allow BigSizedJoinIterator#buildPartitioner to produce more sub-partitions to avoid CudfColumnSizeOverflowException
107+
108+
Note: There is a known issue in the 25.04.0 release when decompressing gzip files on H100 GPUs.
107109
Please find more details in [issue-16661](https://github.com/rapidsai/cudf/issues/16661).
108110

109111
For a detailed list of changes, please refer to the

0 commit comments

Comments
 (0)