Add SC'25 MLP-Offload Artifacts

Author: amaurya

.gitmodules

@@ -0,0 +1,8 @@
+[submodule "mlp-offload/DeepSpeed"]
+	path = mlp-offload/DeepSpeed
+	url = https://github.com/DataStates/DeepSpeed.git
+	branch = dist_nvme_opt
+[submodule "mlp-offload/Megatron-DeepSpeed"]
+	path = mlp-offload/Megatron-DeepSpeed
+	url = https://github.com/mauryaavinash95/Megatron-DeepSpeed
+	branch = dist_nvme_opt

MLP-Offload

@@ -0,0 +1 @@
+Subproject commit e181b47b4b30c0b19e508b27e91a64c07ef9d30a

README.md

@@ -0,0 +1,10 @@
+# DataStates Artifacts
+This repository hosts the software artifacts developed to ensure reproducibility of the novel designs proposed through the DataStates project.
+
+Artifacts are organized by publication and can be found in the following subfolders:
+
+1. [Network Architecture Search](./nas/): Robert Underwood, Meghana Madhyastha, Randal Burns, and Bogdan Nicolae. 2024. EvoStore: Towards Scalable Storage of Evolving Learning Models. In Proceedings of the 33rd International Symposium on High-Performance Parallel and Distributed Computing (HPDC '24). Association for Computing Machinery. Pisa, Italy. https://doi.org/10.1145/3625549.3658679.
+
+2. [Deep Optimizer States](./deep-optimizer-states/): Avinash Maurya, Jie Ye, M. Mustafa Rafique, Franck Cappello, and Bogdan Nicolae. 2024. Deep Optimizer States: Towards Scalable Training of Transformer Models using Interleaved Offloading. In Proceedings of the 25th International Middleware Conference (Middleware '24). Association for Computing Machinery. Hong Kong. https://doi.org/10.1145/3652892.3700781
+
+3. [MLP-Offload](./MLP-Offload/): Avinash Maurya, M. Mustafa Rafique, Franck Cappello, and Bogdan Nicolae. 2025. MLP-Offload: Multi-Level, Multi-Path Offloading for LLM Pre-training to Break the GPU Memory Wall. In Proceedings of the International Conference for High-Performance Computing, Networking, Storage, and Analysis (SC '25). Association for Computing Machinery. St. Lous, MO. https://doi.org/10.1145/3712285.3759864.