Skip to content

[codex] Add embedded DCI simulation backend #13

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
diputs18 wants to merge 2 commits into
base: main
Choose a base branch
from
Open

[codex] Add embedded DCI simulation backend #13

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fa2ad85
Add embedded DCI simulation backend
diputs18 Jun 7, 2026
e516a51
Make topology route follow live task state
diputs18 Jun 7, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
109 changes: 109 additions & 0 deletions configs/dci_runtime_topology.example.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,109 @@
{
"cluster": {
"name": "dci-runtime-topology",
"description": "Long-running DCI simulation topology for embedded or standalone Tianjun demos."
},
"time_scale": 0.08,
"physical_topology": {
"topology_id": "huawei_public_case_reference_dci_runtime_v1",
"topology_nodes": ["USER", "PE1", "PE2", "PE3", "Border1", "Border2", "Border3", "IPCORE", "DC1", "DC2", "DC3"],
"topology_edges": [
{"source": "USER", "target": "PE1", "propagation_delay_ms": 1.8, "bandwidth_mbps": 2000},
{"source": "USER", "target": "PE2", "propagation_delay_ms": 2.2, "bandwidth_mbps": 1800},
{"source": "USER", "target": "PE3", "propagation_delay_ms": 2.0, "bandwidth_mbps": 1900},
{"source": "PE1", "target": "Border1", "propagation_delay_ms": 1.4, "bandwidth_mbps": 4000},
{"source": "PE2", "target": "Border2", "propagation_delay_ms": 1.6, "bandwidth_mbps": 3800},
{"source": "PE3", "target": "Border3", "propagation_delay_ms": 1.5, "bandwidth_mbps": 3900},
{"source": "Border1", "target": "IPCORE", "propagation_delay_ms": 6.2, "bandwidth_mbps": 10000},
{"source": "Border2", "target": "IPCORE", "propagation_delay_ms": 6.5, "bandwidth_mbps": 10000},
{"source": "Border3", "target": "IPCORE", "propagation_delay_ms": 6.1, "bandwidth_mbps": 10000},
{"source": "Border1", "target": "DC1", "propagation_delay_ms": 1.2, "bandwidth_mbps": 8000},
{"source": "Border2", "target": "DC2", "propagation_delay_ms": 1.3, "bandwidth_mbps": 7600},
{"source": "Border3", "target": "DC3", "propagation_delay_ms": 1.2, "bandwidth_mbps": 7800}
],
"compute_attachments": {
"dci-dc1-beijing-vm-0": "DC1",
"dci-dc1-beijing-vm-1": "DC1",
"dci-dc1-beijing-vm-2": "DC1",
"dci-dc1-beijing-vm-3": "DC1",
"dci-dc1-hangzhou-vm-0": "DC1",
"dci-dc1-hangzhou-vm-1": "DC1",
"dci-dc1-hangzhou-vm-2": "DC1",
"dci-dc1-hangzhou-vm-3": "DC1",
"dci-dc2-chengdu-vm-0": "DC2",
"dci-dc2-chengdu-vm-1": "DC2",
"dci-dc2-chengdu-vm-2": "DC2",
"dci-dc2-chengdu-vm-3": "DC2",
"dci-dc2-chongqing-vm-0": "DC2",
"dci-dc2-chongqing-vm-1": "DC2",
"dci-dc2-chongqing-vm-2": "DC2",
"dci-dc2-chongqing-vm-3": "DC2",
"dci-dc3-guangzhou-vm-0": "DC3",
"dci-dc3-guangzhou-vm-1": "DC3",
"dci-dc3-guangzhou-vm-2": "DC3",
"dci-dc3-guangzhou-vm-3": "DC3",
"dci-dc3-shenzhen-vm-0": "DC3",
"dci-dc3-shenzhen-vm-1": "DC3",
"dci-dc3-shenzhen-vm-2": "DC3",
"dci-dc3-shenzhen-vm-3": "DC3"
},
"provenance": {
"source": "Tianjun embedded runtime demo",
"purpose": "persistent local topology that can consume Hermes pending tasks through leases"
}
},
"links": [
{"source_region": "east", "target_region": "dc1", "latency_ms": {"p50": 2.4, "jitter": 0.5}, "bandwidth_mbps": {"p50": 3600, "jitter": 160}, "packet_loss": 0.0005, "path_reliability": 0.998},
{"source_region": "east", "target_region": "dc2", "latency_ms": {"p50": 16.3, "jitter": 2.1}, "bandwidth_mbps": {"p50": 2400, "jitter": 260}, "packet_loss": 0.002, "path_reliability": 0.994},
{"source_region": "east", "target_region": "dc3", "latency_ms": {"p50": 16.1, "jitter": 1.9}, "bandwidth_mbps": {"p50": 2600, "jitter": 240}, "packet_loss": 0.0018, "path_reliability": 0.995},
{"source_region": "west", "target_region": "dc1", "latency_ms": {"p50": 16.3, "jitter": 2.2}, "bandwidth_mbps": {"p50": 2300, "jitter": 280}, "packet_loss": 0.0022, "path_reliability": 0.993},
{"source_region": "west", "target_region": "dc2", "latency_ms": {"p50": 2.6, "jitter": 0.6}, "bandwidth_mbps": {"p50": 3500, "jitter": 170}, "packet_loss": 0.0006, "path_reliability": 0.998},
{"source_region": "west", "target_region": "dc3", "latency_ms": {"p50": 17.0, "jitter": 2.4}, "bandwidth_mbps": {"p50": 2200, "jitter": 300}, "packet_loss": 0.0024, "path_reliability": 0.992},
{"source_region": "south", "target_region": "dc1", "latency_ms": {"p50": 16.1, "jitter": 1.9}, "bandwidth_mbps": {"p50": 2600, "jitter": 240}, "packet_loss": 0.0018, "path_reliability": 0.995},
{"source_region": "south", "target_region": "dc2", "latency_ms": {"p50": 17.0, "jitter": 2.4}, "bandwidth_mbps": {"p50": 2200, "jitter": 300}, "packet_loss": 0.0024, "path_reliability": 0.992},
{"source_region": "south", "target_region": "dc3", "latency_ms": {"p50": 2.5, "jitter": 0.5}, "bandwidth_mbps": {"p50": 3550, "jitter": 165}, "packet_loss": 0.0005, "path_reliability": 0.998}
],
"workload_profiles": {
"batch": {
"execution_time": {"base_seconds": 2, "per_cpu_second": 0.45, "per_memory_gb_second": 0.03, "per_input_gb_second": 0.2, "minimum_seconds": 2, "maximum_seconds": 36},
"capacity": {"qps_per_cpu": 10, "qps_per_gpu": 0},
"pricing": {"on_demand_per_hour": 8, "spot_per_hour": 4}
},
"batch_cpu": {
"execution_time": {"base_seconds": 2, "per_cpu_second": 0.45, "per_memory_gb_second": 0.03, "per_input_gb_second": 0.2, "minimum_seconds": 2, "maximum_seconds": 36},
"capacity": {"qps_per_cpu": 10, "qps_per_gpu": 0},
"pricing": {"on_demand_per_hour": 8, "spot_per_hour": 4}
},
"inference": {
"execution_time": {"base_seconds": 3, "per_cpu_second": 0.3, "per_gpu_second": 0.8, "per_memory_gb_second": 0.02, "minimum_seconds": 3, "maximum_seconds": 45},
"capacity": {"qps_per_gpu": 120, "qps_per_cpu": 12, "max_batch_size": 8},
"pricing": {"on_demand_per_hour": 18, "spot_per_hour": 9}
}
},
"nodes": [
{"node_id": "dci-dc1-beijing-vm-0", "region": "dc1", "location": "beijing", "service_region": "east", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "latency-sensitive"], "capacity": {"cpu": 8, "memory": 32, "gpu": 0, "storage": 120}, "cost_per_tick": 0.9, "base_reliability": 0.985, "health_score": 0.94, "performance_factors": {"batch": 1.12, "batch_cpu": 1.12, "inference": 0.95}},
{"node_id": "dci-dc1-beijing-vm-1", "region": "dc1", "location": "beijing", "service_region": "east", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "latency-sensitive"], "capacity": {"cpu": 8, "memory": 32, "gpu": 0, "storage": 120}, "cost_per_tick": 0.92, "base_reliability": 0.984, "health_score": 0.935, "performance_factors": {"batch": 1.08, "batch_cpu": 1.08, "inference": 0.95}},
{"node_id": "dci-dc1-beijing-vm-2", "region": "dc1", "location": "beijing", "service_region": "east", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 160}, "cost_per_tick": 1.05, "base_reliability": 0.982, "health_score": 0.91, "performance_factors": {"batch": 1.18, "batch_cpu": 1.18, "inference": 1.0}},
{"node_id": "dci-dc1-beijing-vm-3", "region": "dc1", "location": "beijing", "service_region": "east", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 160}, "cost_per_tick": 1.06, "base_reliability": 0.981, "health_score": 0.91, "performance_factors": {"batch": 1.15, "batch_cpu": 1.15, "inference": 1.0}},
{"node_id": "dci-dc1-hangzhou-vm-0", "region": "dc1", "location": "hangzhou", "service_region": "east", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "low-latency"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 200}, "cost_per_tick": 1.12, "base_reliability": 0.983, "health_score": 0.891, "performance_factors": {"batch": 1.24, "batch_cpu": 1.24, "inference": 1.05}},
{"node_id": "dci-dc1-hangzhou-vm-1", "region": "dc1", "location": "hangzhou", "service_region": "east", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "low-latency"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 200}, "cost_per_tick": 1.1, "base_reliability": 0.984, "health_score": 0.899, "performance_factors": {"batch": 1.28, "batch_cpu": 1.28, "inference": 1.05}},
{"node_id": "dci-dc1-hangzhou-vm-2", "region": "dc1", "location": "hangzhou", "service_region": "east", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "low-latency"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 1.0, "base_reliability": 0.982, "health_score": 0.889, "performance_factors": {"batch": 1.2, "batch_cpu": 1.2, "inference": 1.02}},
{"node_id": "dci-dc1-hangzhou-vm-3", "region": "dc1", "location": "hangzhou", "service_region": "east", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "low-latency"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 1.01, "base_reliability": 0.981, "health_score": 0.877, "performance_factors": {"batch": 1.16, "batch_cpu": 1.16, "inference": 1.0}},
{"node_id": "dci-dc2-chengdu-vm-0", "region": "dc2", "location": "chengdu", "service_region": "west", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 0.96, "base_reliability": 0.982, "health_score": 0.902, "performance_factors": {"batch": 1.1, "batch_cpu": 1.1, "inference": 0.98}},
{"node_id": "dci-dc2-chengdu-vm-1", "region": "dc2", "location": "chengdu", "service_region": "west", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 0.95, "base_reliability": 0.983, "health_score": 0.922, "performance_factors": {"batch": 1.12, "batch_cpu": 1.12, "inference": 0.98}},
{"node_id": "dci-dc2-chengdu-vm-2", "region": "dc2", "location": "chengdu", "service_region": "west", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 220}, "cost_per_tick": 1.08, "base_reliability": 0.984, "health_score": 0.926, "performance_factors": {"batch": 1.2, "batch_cpu": 1.2, "inference": 1.0}},
{"node_id": "dci-dc2-chengdu-vm-3", "region": "dc2", "location": "chengdu", "service_region": "west", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 220}, "cost_per_tick": 1.09, "base_reliability": 0.985, "health_score": 0.944, "performance_factors": {"batch": 1.22, "batch_cpu": 1.22, "inference": 1.0}},
{"node_id": "dci-dc2-chongqing-vm-0", "region": "dc2", "location": "chongqing", "service_region": "west", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "reliable"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 220}, "cost_per_tick": 1.07, "base_reliability": 0.986, "health_score": 0.953, "performance_factors": {"batch": 1.2, "batch_cpu": 1.2, "inference": 1.0}},
{"node_id": "dci-dc2-chongqing-vm-1", "region": "dc2", "location": "chongqing", "service_region": "west", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "reliable"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 220}, "cost_per_tick": 1.07, "base_reliability": 0.987, "health_score": 0.959, "performance_factors": {"batch": 1.22, "batch_cpu": 1.22, "inference": 1.0}},
{"node_id": "dci-dc2-chongqing-vm-2", "region": "dc2", "location": "chongqing", "service_region": "west", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "reliable"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 0.98, "base_reliability": 0.986, "health_score": 0.957, "performance_factors": {"batch": 1.14, "batch_cpu": 1.14, "inference": 0.98}},
{"node_id": "dci-dc2-chongqing-vm-3", "region": "dc2", "location": "chongqing", "service_region": "west", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "reliable"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 0.99, "base_reliability": 0.987, "health_score": 0.96, "performance_factors": {"batch": 1.16, "batch_cpu": 1.16, "inference": 0.98}},
{"node_id": "dci-dc3-guangzhou-vm-0", "region": "dc3", "location": "guangzhou", "service_region": "south", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 220}, "cost_per_tick": 1.05, "base_reliability": 0.985, "health_score": 0.95, "performance_factors": {"batch": 1.18, "batch_cpu": 1.18, "inference": 1.0}},
{"node_id": "dci-dc3-guangzhou-vm-1", "region": "dc3", "location": "guangzhou", "service_region": "south", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 220}, "cost_per_tick": 1.05, "base_reliability": 0.984, "health_score": 0.947, "performance_factors": {"batch": 1.16, "batch_cpu": 1.16, "inference": 1.0}},
{"node_id": "dci-dc3-guangzhou-vm-2", "region": "dc3", "location": "guangzhou", "service_region": "south", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 0.97, "base_reliability": 0.982, "health_score": 0.93, "performance_factors": {"batch": 1.1, "batch_cpu": 1.1, "inference": 0.98}},
{"node_id": "dci-dc3-guangzhou-vm-3", "region": "dc3", "location": "guangzhou", "service_region": "south", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 0.98, "base_reliability": 0.981, "health_score": 0.917, "performance_factors": {"batch": 1.08, "batch_cpu": 1.08, "inference": 0.98}},
{"node_id": "dci-dc3-shenzhen-vm-0", "region": "dc3", "location": "shenzhen", "service_region": "south", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "low-latency"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 220}, "cost_per_tick": 1.08, "base_reliability": 0.983, "health_score": 0.905, "performance_factors": {"batch": 1.2, "batch_cpu": 1.2, "inference": 1.02}},
{"node_id": "dci-dc3-shenzhen-vm-1", "region": "dc3", "location": "shenzhen", "service_region": "south", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "low-latency"], "capacity": {"cpu": 16, "memory": 64, "gpu": 0, "storage": 220}, "cost_per_tick": 1.08, "base_reliability": 0.982, "health_score": 0.895, "performance_factors": {"batch": 1.18, "batch_cpu": 1.18, "inference": 1.02}},
{"node_id": "dci-dc3-shenzhen-vm-2", "region": "dc3", "location": "shenzhen", "service_region": "south", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "low-latency"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 1.0, "base_reliability": 0.981, "health_score": 0.888, "performance_factors": {"batch": 1.1, "batch_cpu": 1.1, "inference": 1.0}},
{"node_id": "dci-dc3-shenzhen-vm-3", "region": "dc3", "location": "shenzhen", "service_region": "south", "labels": ["dci", "cloudsim", "cloudsimplus", "cpu", "low-latency"], "capacity": {"cpu": 12, "memory": 48, "gpu": 0, "storage": 180}, "cost_per_tick": 1.01, "base_reliability": 0.981, "health_score": 0.886, "performance_factors": {"batch": 1.08, "batch_cpu": 1.08, "inference": 1.0}}
]
}
8 changes: 8 additions & 0 deletions configs/tianjun.example.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,14 @@ server = "http://127.0.0.1:8024"
max_cycles = 30
poll_interval_seconds = 1.0

[simulation]
server = "http://127.0.0.1:8024"
inventory = "${TIANJUN_HOME}/configs/dci_runtime_topology.example.json"
auto_start = false
poll_interval_seconds = 1.0
time_scale = 0.08
verbose = false

[real_agent]
server = "http://127.0.0.1:8024"
node_config = "${TIANJUN_HOME}/examples/real_node_agent.example.json"
Expand Down
7 changes: 3 additions & 4 deletions restart_tianjun.bat
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,8 +37,8 @@ if errorlevel 1 (

echo Starting Tianjun Engine control plane...
echo Dashboard: %URL%
echo Starting control plane...
start "Tianjun Control Plane" cmd /k python -B main.py serve --config "%CONFIG%" --default-execution-mode simulation --host %HOST% --port %PORT%
echo Starting control plane with embedded DCI simulation backend...
start "Tianjun Control Plane" cmd /k python -B main.py serve --config "%CONFIG%" --default-execution-mode simulation --host %HOST% --port %PORT% --auto-sim-backend

echo Waiting for control plane health check...
powershell -NoProfile -ExecutionPolicy Bypass -Command ^
Expand All @@ -52,6 +52,5 @@ if errorlevel 1 (
start "" "%URL%"

echo Tianjun Engine control plane is running in a separate window.
echo No simulated nodes are started automatically.
echo Start CloudSim Plus, sim-backend, or a node agent manually when you want nodes to appear.
echo Embedded DCI simulated nodes are running and will pull pending tasks automatically.
pause
Loading