Skip to main content
Glama

Physics MCP Server

by BlinkZer0
phase7_distributed_collaboration.md4.54 kB
# Phase 7: Distributed Collaboration Examples This document provides examples of using the `distributed_collaboration` tool for distributed computing and collaboration features. ## Job Submit Examples ### Slurm Job Submission ```json { "tool": "distributed_collaboration", "method": "job_submit", "backend": "slurm", "job_spec": { "resources": { "cpu": 4, "memory": "8GB", "gpu": 1, "time_limit": "2:00:00" }, "command": ["python", "physics_simulation.py"], "env": { "CUDA_VISIBLE_DEVICES": "0", "OMP_NUM_THREADS": "4" } }, "artifacts_path": "/scratch/user/results", "stream_logs": true, "timeout_sec": 7200 } ``` ### Kubernetes Job Submission ```json { "tool": "distributed_collaboration", "method": "job_submit", "backend": "k8s", "job_spec": { "resources": { "cpu": 2, "memory": "4Gi", "gpu": 0 }, "image": "python:3.11-slim", "command": ["python", "-c", "import numpy as np; print('Hello from K8s!')"], "env": { "PYTHONPATH": "/app" }, "mounts": [ { "source": "/data", "target": "/app/data", "readonly": true } ] }, "artifacts_path": "/app/output", "stream_logs": true } ``` ## Session Share Examples ### Create Read-Only Share ```json { "tool": "distributed_collaboration", "method": "session_share", "session_id": "session_abc123", "access": "read", "expires_in_hours": 72, "participants": ["alice@example.com", "bob@example.com"] } ``` ### Create Collaborative Share ```json { "tool": "distributed_collaboration", "method": "session_share", "session_id": "session_def456", "access": "write", "expires_in_hours": 168, "participants": ["team@physics-lab.edu"] } ``` ## Lab Notebook Examples ### Create Notebook Entry ```json { "tool": "distributed_collaboration", "method": "lab_notebook", "session_id": "session_abc123", "title": "Hydrogen 2p Orbital Analysis", "notes_md": "# Analysis Results\n\nSuccessfully computed the 2p orbital wavefunction with high precision.\n\n## Key Findings\n- Energy eigenvalue: -3.4 eV\n- Radial nodes: 0\n- Angular momentum: l=1\n\n## Next Steps\n- Compare with experimental data\n- Extend to 3d orbitals", "attach_artifacts": [ "artifacts/session_abc123/wavefunction_plot.png", "artifacts/session_abc123/energy_data.csv" ], "sign_as": "Dr. Jane Smith" } ``` ## Artifact Versioning Examples ### Register Artifacts with Lineage ```json { "tool": "distributed_collaboration", "method": "artifact_versioning", "artifacts": [ "artifacts/session_abc123/final_plot.png", "artifacts/session_abc123/simulation_data.csv", "artifacts/session_abc123/analysis_report.pdf" ], "parents": [ "sha256:abc123def456...", "sha256:789xyz012..." ], "params_json": { "temperature": 300, "pressure": 1.0, "method": "DFT", "basis_set": "6-31G*" }, "code_version": "v2.1.3" } ``` ## Legacy Tool Usage The individual methods can also be called directly for backward compatibility: ```json { "tool": "job_submit", "backend": "slurm", "job_spec": {...}, "artifacts_path": "/scratch/results" } ``` ```json { "tool": "session_share", "session_id": "session_123", "access": "read" } ``` ## Response Examples ### Job Submit Response ```json { "job_id": "slurm_12345", "log_stream_path": "artifacts/session_abc123/job_12345.log", "returned_artifacts": [ "artifacts/session_abc123/result_plot.png", "artifacts/session_abc123/output_data.csv" ], "meta": { "backend": "slurm", "device": "cuda", "mesh": [64, 64], "commit": "a1b2c3d4", "duration_ms": 45000 } } ``` ### Session Share Response ```json { "share_url": "https://phys-mcp.example.com/share/abc123def456", "expires_at": "2024-01-15T12:00:00Z", "participants": ["alice@example.com", "bob@example.com"] } ``` ### Lab Notebook Response ```json { "entry_id": "notebook_789xyz", "pdf_path": "artifacts/session_abc123/notebook_789xyz.pdf", "meta": { "hash": "sha256:def456abc789..." } } ``` ### Artifact Versioning Response ```json { "records": [ { "artifact": "artifacts/session_abc123/final_plot.png", "hash": "sha256:abc123def456...", "lineage_id": "lineage_xyz789" }, { "artifact": "artifacts/session_abc123/simulation_data.csv", "hash": "sha256:def456abc123...", "lineage_id": "lineage_abc456" } ], "meta": { "cached": false } } ```

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/BlinkZer0/Phys-MCP'

If you have feedback or need assistance with the MCP directory API, please join our Discord server