Dropping cuGraph Service (#5325)

Dropping cuGraph service


Replaces #5315

Authors:
  - Brad Rees (https://github.com/BradReesWork)

Approvers:
  - Alex Barghi (https://github.com/alexbarghi-nv)
  - Bradley Dice (https://github.com/bdice)
  - Don Acosta (https://github.com/acostadon)

URL: #5325

Author: BradReesWork

README.md

@@ -35,13 +35,12 @@ The [cugraph-docs repository](https://github.com/rapidsai/cugraph-docs) contains
 
 <br>
 
-[RAPIDS](https://rapids.ai) cuGraph is a repo that represents a collection of packages focused on GPU-accelerated graph analytics including support for property graphs and remote (graph as a service) operations.  cuGraph supports the creation and manipulation of graphs followed by the execution of scalable fast graph algorithms.
+[RAPIDS](https://rapids.ai) cuGraph is a repo that represents a collection of packages focused on GPU-accelerated graph analytics including support for property graphs.  cuGraph supports the creation and manipulation of graphs followed by the execution of scalable fast graph algorithms.
 
 <div align="center">
 
 [Getting cuGraph](https://docs.rapids.ai/api/cugraph/nightly/) *
 [Graph Algorithms](https://docs.rapids.ai/api/cugraph/nightly/graph_support/algorithms/) *
-[Graph Service](./readme_pages/cugraph_service.md) *
 [Property Graph](./readme_pages/property_graph.md) *
 
 </div>
@@ -64,7 +63,6 @@ The [cugraph-docs repository](https://github.com/rapidsai/cugraph-docs) contains
   - [pylibcugraph](./readme_pages/pylibcugraph.md)
   - [libcugraph (C/C++/CUDA)](./readme_pages/libcugraph.md)
   - [nx-cugraph](https://rapids.ai/nx-cugraph/)
-  - [cugraph-service](./readme_pages/cugraph_service.md)
 - API Docs
   - Python
     - [Python Nightly](https://docs.rapids.ai/api/cugraph/nightly/api_docs/cugraph/)
@@ -86,7 +84,7 @@ The [cugraph-docs repository](https://github.com/rapidsai/cugraph-docs) contains
 
 <img src="img/Stack2.png" alt="Stack" width="800">
 
-[RAPIDS](https://rapids.ai) cuGraph is a collection of GPU-accelerated graph algorithms and services. At the Python layer, cuGraph operates on [GPU DataFrames](https://github.com/rapidsai/cudf), thereby allowing for seamless passing of data between ETL tasks in [cuDF](https://github.com/rapidsai/cudf) and machine learning tasks in [cuML](https://github.com/rapidsai/cuml). Data scientists familiar with Python will quickly pick up how cuGraph integrates with the Pandas-like API of cuDF.  Likewise, users familiar with NetworkX will quickly recognize the NetworkX-like API provided in cuGraph, with the goal to allow existing code to be ported with minimal effort into RAPIDS. To simplify integration, cuGraph also supports data found in [Pandas DataFrame](https://pandas.pydata.org/), [NetworkX Graph Objects](https://networkx.org/) and several other formats.
+[RAPIDS](https://rapids.ai) cuGraph is a collection of GPU-accelerated graph algorithms. At the Python layer, cuGraph operates on [GPU DataFrames](https://github.com/rapidsai/cudf), thereby allowing for seamless passing of data between ETL tasks in [cuDF](https://github.com/rapidsai/cudf) and machine learning tasks in [cuML](https://github.com/rapidsai/cuml). Data scientists familiar with Python will quickly pick up how cuGraph integrates with the Pandas-like API of cuDF.  Likewise, users familiar with NetworkX will quickly recognize the NetworkX-like API provided in cuGraph, with the goal to allow existing code to be ported with minimal effort into RAPIDS. To simplify integration, cuGraph also supports data found in [Pandas DataFrame](https://pandas.pydata.org/), [NetworkX Graph Objects](https://networkx.org/) and several other formats.
 
 While the high-level cugraph python API provides an easy-to-use and familiar interface for data scientists that's consistent with other RAPIDS libraries in their workflow, some use cases require access to lower-level graph theory concepts.  For these users, we provide an additional Python API called pylibcugraph, intended for applications that require a tighter integration with cuGraph at the Python layer with fewer dependencies.  Users familiar with C/C++/CUDA and graph structures can access libcugraph and libcugraph_c for low level integration outside of python.
 

benchmarks/cugraph-service/pytest-based/bench_cgs_client_scaling.py

@@ -7,7 +7,6 @@ pythonpath =
 
 testpaths =
     cugraph/pytest_based
-    cugraph-service/pytest_based
 
 addopts =
     --benchmark-columns="min, max, mean, stddev, outliers"
@@ -32,7 +31,6 @@ markers =
     snmg: single-node multi-GPU
     mnmg: multi-node multi-GPU
     local: local cugraph
-    remote: cugraph-service
     batch_size_100: batch size of 100 for sampling algos
     batch_size_500: batch size of 500 for sampling algos
     batch_size_1000: batch size of 1000 for sampling algos
@@ -48,11 +46,6 @@ markers =
     batch_size_80000: batch size of 80000 for sampling algos
     batch_size_90000: batch size of 90000 for sampling algos
     batch_size_100000: batch size of 100000 for sampling algos
-    num_clients_2: start 2 cugraph-service clients
-    num_clients_4: start 4 cugraph-service clients
-    num_clients_8: start 8 cugraph-service clients
-    num_clients_16: start 16 cugraph-service clients
-    num_clients_32: start 32 cugraph-service clients
     fanout_10_25: fanout [10, 25] for sampling algos
     fanout_5_10_15: fanout [5, 10, 15] for sampling algos
     rmat_data: RMAT-generated synthetic datasets

benchmarks/cugraph-service/pytest-based/bench_cgs_uniform_neighbor_sample.py

@@ -1,4 +1,4 @@
-# SPDX-FileCopyrightText: Copyright (c) 2022-2024, NVIDIA CORPORATION.
+# SPDX-FileCopyrightText: Copyright (c) 2022-2025, NVIDIA CORPORATION.
 # SPDX-License-Identifier: Apache-2.0
 
 import pytest
@@ -165,16 +165,6 @@
     id="fanout=5_10_15",
 )
 
-# scaling the number of concurrent cugraph-service clients
-_num_clients = [2, 4, 8, 16, 32]
-num_clients = {}
-for nc in _num_clients:
-    num_clients[nc] = pytest.param(
-        nc,
-        id=f"num_clients={nc}",
-        marks=[getattr(pytest.mark, f"num_clients_{nc}")],
-    )
-
 # Parameters for Graph generation fixture
 graph_obj_fixture_params = gen_fixture_params(
     (sg, rmat["20_16"]),