torc 0.21.0

Workflow management system
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156

//! LLNL Dane HPC profile
//!
//! Dane is a cluster at Lawrence Livermore National Laboratory (LLNL) featuring:
//! - CPU nodes with 112 cores and ~251GB RAM each
//! - Various partitions for different walltime requirements
//!
//! Detection: Environment variable CLUSTER=dane

use super::profiles::{HpcDetection, HpcPartition, HpcProfile};

/// Create the Dane HPC profile
pub fn dane_profile() -> HpcProfile {
    HpcProfile {
        name: "dane".to_string(),
        display_name: "LLNL Dane".to_string(),
        description: "Lawrence Livermore National Laboratory's Dane cluster".to_string(),
        detection: vec![HpcDetection::EnvVar {
            name: "CLUSTER".to_string(),
            value: "dane".to_string(),
        }],
        default_account: None,
        partitions: dane_partitions(),
        charge_factor_cpu: 1.0,
        charge_factor_gpu: 1.0, // No GPUs on this system
        metadata: [
            (
                "documentation".to_string(),
                "https://hpc.llnl.gov/hardware/compute-platforms/dane".to_string(),
            ),
            ("organization".to_string(), "LLNL".to_string()),
        ]
        .into_iter()
        .collect(),
    }
}

fn dane_partitions() -> Vec<HpcPartition> {
    vec![
        // Debug partition - short walltime for testing
        HpcPartition {
            name: "pdebug".to_string(),
            description: "Debug partition for developing and testing jobs".to_string(),
            cpus_per_node: 112,
            memory_mb: 257_054,
            max_walltime_secs: 3600, // 1 hour
            max_nodes: None,
            max_nodes_per_user: None,
            min_nodes: None,
            gpus_per_node: None,
            gpu_type: None,
            gpu_memory_gb: None,
            local_disk_gb: None,
            shared: false,
            requires_explicit_request: true,
            default_qos: None,
            features: vec!["debug".to_string()],
        },
        // Default batch partition (1 day max)
        HpcPartition {
            name: "pbatch".to_string(),
            description: "Default batch partition for standard jobs (max 1 day)".to_string(),
            cpus_per_node: 112,
            memory_mb: 257_054,
            max_walltime_secs: 24 * 3600, // 1 day
            max_nodes: None,
            max_nodes_per_user: None,
            min_nodes: None,
            gpus_per_node: None,
            gpu_type: None,
            gpu_memory_gb: None,
            local_disk_gb: None,
            shared: false,
            requires_explicit_request: false, // Default partition
            default_qos: None,
            features: vec![],
        },
        // CI partition (1 day max)
        HpcPartition {
            name: "pci".to_string(),
            description: "CI partition for continuous integration jobs".to_string(),
            cpus_per_node: 112,
            memory_mb: 257_054,
            max_walltime_secs: 24 * 3600, // 1 day
            max_nodes: None,
            max_nodes_per_user: None,
            min_nodes: None,
            gpus_per_node: None,
            gpu_type: None,
            gpu_memory_gb: None,
            local_disk_gb: None,
            shared: false,
            requires_explicit_request: true,
            default_qos: None,
            features: vec!["ci".to_string()],
        },
        // Serial partition - long walltime for serial jobs
        HpcPartition {
            name: "pserial".to_string(),
            description: "Serial partition for long-running jobs (max 7 days)".to_string(),
            cpus_per_node: 112,
            memory_mb: 257_054,
            max_walltime_secs: 7 * 24 * 3600, // 7 days
            max_nodes: None,
            max_nodes_per_user: None,
            min_nodes: None,
            gpus_per_node: None,
            gpu_type: None,
            gpu_memory_gb: None,
            local_disk_gb: None,
            shared: false,
            requires_explicit_request: true,
            default_qos: None,
            features: vec!["serial".to_string()],
        },
        // Jupyter partition
        HpcPartition {
            name: "pjupyter".to_string(),
            description: "Partition for Jupyter notebook sessions".to_string(),
            cpus_per_node: 112,
            memory_mb: 257_054,
            max_walltime_secs: 24 * 3600, // 1 day
            max_nodes: None,
            max_nodes_per_user: None,
            min_nodes: None,
            gpus_per_node: None,
            gpu_type: None,
            gpu_memory_gb: None,
            local_disk_gb: None,
            shared: true, // Jupyter sessions typically share nodes
            requires_explicit_request: true,
            default_qos: None,
            features: vec!["jupyter".to_string()],
        },
        // All partition - special access, infinite walltime
        // Note: This partition likely requires special permissions
        HpcPartition {
            name: "pall".to_string(),
            description: "Special partition with unlimited walltime (requires special access)"
                .to_string(),
            cpus_per_node: 112,
            memory_mb: 257_054,
            max_walltime_secs: 365 * 24 * 3600, // Effectively infinite (1 year)
            max_nodes: None,
            max_nodes_per_user: None,
            min_nodes: None,
            gpus_per_node: None,
            gpu_type: None,
            gpu_memory_gb: None,
            local_disk_gb: None,
            shared: false,
            requires_explicit_request: true,
            default_qos: None,
            features: vec!["special".to_string()],
        },
    ]
}