2022-06-22 11:20:57 +02:00
|
|
|
<script>
|
|
|
|
import Refresher from './joblist/Refresher.svelte'
|
|
|
|
import Roofline, { transformPerNodeData } from './plots/Roofline.svelte'
|
|
|
|
import Histogram from './plots/Histogram.svelte'
|
2022-09-28 16:13:46 +02:00
|
|
|
import { Row, Col, Spinner, Card, CardHeader, CardTitle, CardBody, Table, Progress, Icon } from 'sveltestrap'
|
2023-04-12 18:00:28 +02:00
|
|
|
import { init } from './utils.js'
|
2023-06-16 12:44:34 +02:00
|
|
|
import { scaleNumbers } from './units.js'
|
2023-05-03 16:41:17 +02:00
|
|
|
import { queryStore, gql, getContextClient } from '@urql/svelte'
|
2022-06-22 11:20:57 +02:00
|
|
|
|
|
|
|
const { query: initq } = init()
|
|
|
|
|
|
|
|
export let cluster
|
|
|
|
|
|
|
|
let plotWidths = [], colWidth1 = 0, colWidth2
|
|
|
|
let from = new Date(Date.now() - 5 * 60 * 1000), to = new Date(Date.now())
|
2023-05-10 16:35:21 +02:00
|
|
|
|
2023-05-12 11:19:37 +02:00
|
|
|
const client = getContextClient();
|
2023-05-10 16:35:21 +02:00
|
|
|
$: mainQuery = queryStore({
|
2023-05-12 11:19:37 +02:00
|
|
|
client: client,
|
2023-05-03 16:41:17 +02:00
|
|
|
query: gql`query($cluster: String!, $filter: [JobFilter!]!, $metrics: [String!], $from: Time!, $to: Time!) {
|
2022-06-22 11:20:57 +02:00
|
|
|
nodeMetrics(cluster: $cluster, metrics: $metrics, from: $from, to: $to) {
|
2023-03-30 15:21:35 +02:00
|
|
|
host
|
|
|
|
subCluster
|
2022-06-22 11:20:57 +02:00
|
|
|
metrics {
|
2023-03-30 15:21:35 +02:00
|
|
|
name
|
|
|
|
scope
|
2022-06-22 11:20:57 +02:00
|
|
|
metric {
|
2023-03-30 15:21:35 +02:00
|
|
|
timestep
|
|
|
|
unit { base, prefix }
|
2022-06-22 11:20:57 +02:00
|
|
|
series { data }
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
stats: jobsStatistics(filter: $filter) {
|
|
|
|
histDuration { count, value }
|
|
|
|
histNumNodes { count, value }
|
|
|
|
}
|
|
|
|
|
|
|
|
allocatedNodes(cluster: $cluster) { name, count }
|
|
|
|
topUsers: jobsCount(filter: $filter, groupBy: USER, weight: NODE_COUNT, limit: 10) { name, count }
|
|
|
|
topProjects: jobsCount(filter: $filter, groupBy: PROJECT, weight: NODE_COUNT, limit: 10) { name, count }
|
2023-05-03 16:41:17 +02:00
|
|
|
}`,
|
|
|
|
variables: {
|
|
|
|
cluster: cluster, metrics: ['flops_any', 'mem_bw'], from: from.toISOString(), to: to.toISOString(),
|
2022-06-22 11:20:57 +02:00
|
|
|
filter: [{ state: ['running'] }, { cluster: { eq: cluster } }]
|
2023-05-03 16:41:17 +02:00
|
|
|
}
|
2022-06-22 11:20:57 +02:00
|
|
|
})
|
|
|
|
|
|
|
|
const sumUp = (data, subcluster, metric) => data.reduce((sum, node) => node.subCluster == subcluster
|
|
|
|
? sum + (node.metrics.find(m => m.name == metric)?.metric.series.reduce((sum, series) => sum + series.data[series.data.length - 1], 0) || 0)
|
|
|
|
: sum, 0)
|
|
|
|
|
2023-06-16 12:44:34 +02:00
|
|
|
let allocatedNodes = {}, flopRate = {}, flopRateUnitPrefix = {}, flopRateUnitBase = {}, memBwRate = {}, memBwRateUnitPrefix = {}, memBwRateUnitBase = {}
|
2022-06-22 11:20:57 +02:00
|
|
|
$: if ($initq.data && $mainQuery.data) {
|
|
|
|
let subClusters = $initq.data.clusters.find(c => c.name == cluster).subClusters
|
|
|
|
for (let subCluster of subClusters) {
|
2023-06-16 12:44:34 +02:00
|
|
|
allocatedNodes[subCluster.name] = $mainQuery.data.allocatedNodes.find(({ name }) => name == subCluster.name)?.count || 0
|
|
|
|
flopRate[subCluster.name] = Math.floor(sumUp($mainQuery.data.nodeMetrics, subCluster.name, 'flops_any') * 100) / 100
|
|
|
|
flopRateUnitPrefix[subCluster.name] = subCluster.flopRateSimd.unit.prefix
|
|
|
|
flopRateUnitBase[subCluster.name] = subCluster.flopRateSimd.unit.base
|
|
|
|
memBwRate[subCluster.name] = Math.floor(sumUp($mainQuery.data.nodeMetrics, subCluster.name, 'mem_bw') * 100) / 100
|
|
|
|
memBwRateUnitPrefix[subCluster.name] = subCluster.memoryBandwidth.unit.prefix
|
|
|
|
memBwRateUnitBase[subCluster.name] = subCluster.memoryBandwidth.unit.base
|
2022-06-22 11:20:57 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
</script>
|
|
|
|
|
2022-09-28 16:13:46 +02:00
|
|
|
<!-- Loading indicator & Refresh -->
|
|
|
|
|
2022-06-22 11:20:57 +02:00
|
|
|
<Row>
|
2022-09-28 16:13:46 +02:00
|
|
|
<Col xs="auto" style="align-self: flex-end;">
|
2023-04-12 18:00:28 +02:00
|
|
|
<h4 class="mb-0" >Current utilization of cluster "{cluster}"</h4>
|
2022-09-28 16:13:46 +02:00
|
|
|
</Col>
|
2022-06-22 11:20:57 +02:00
|
|
|
<Col xs="auto">
|
|
|
|
{#if $initq.fetching || $mainQuery.fetching}
|
|
|
|
<Spinner/>
|
|
|
|
{:else if $initq.error}
|
|
|
|
<Card body color="danger">{$initq.error.message}</Card>
|
|
|
|
{:else}
|
|
|
|
<!-- ... -->
|
|
|
|
{/if}
|
|
|
|
</Col>
|
|
|
|
<Col xs="auto" style="margin-left: auto;">
|
|
|
|
<Refresher initially={120} on:reload={() => {
|
|
|
|
from = new Date(Date.now() - 5 * 60 * 1000)
|
|
|
|
to = new Date(Date.now())
|
|
|
|
}} />
|
|
|
|
</Col>
|
|
|
|
</Row>
|
|
|
|
{#if $mainQuery.error}
|
|
|
|
<Row>
|
|
|
|
<Col>
|
|
|
|
<Card body color="danger">{$mainQuery.error.message}</Card>
|
|
|
|
</Col>
|
|
|
|
</Row>
|
|
|
|
{/if}
|
2022-09-28 16:13:46 +02:00
|
|
|
|
|
|
|
<hr>
|
|
|
|
|
|
|
|
<!-- Gauges & Roofline per Subcluster-->
|
|
|
|
|
2022-06-22 11:20:57 +02:00
|
|
|
{#if $initq.data && $mainQuery.data}
|
|
|
|
{#each $initq.data.clusters.find(c => c.name == cluster).subClusters as subCluster, i}
|
2022-09-28 16:13:46 +02:00
|
|
|
<Row cols={2} class="mb-3 justify-content-center">
|
2022-10-07 15:01:14 +02:00
|
|
|
<Col xs="4" class="px-3">
|
2022-09-28 16:13:46 +02:00
|
|
|
<Card class="h-auto mt-1">
|
|
|
|
<CardHeader>
|
|
|
|
<CardTitle class="mb-0">SubCluster "{subCluster.name}"</CardTitle>
|
|
|
|
</CardHeader>
|
|
|
|
<CardBody>
|
2023-04-12 18:00:28 +02:00
|
|
|
<Table borderless>
|
|
|
|
<tr class="py-2">
|
2022-09-28 16:13:46 +02:00
|
|
|
<th scope="col">Allocated Nodes</th>
|
2022-10-07 15:01:14 +02:00
|
|
|
<td style="min-width: 100px;"><div class="col"><Progress value={allocatedNodes[subCluster.name]} max={subCluster.numberOfNodes}/></div></td>
|
2023-06-16 13:01:41 +02:00
|
|
|
<td>{allocatedNodes[subCluster.name]} / {subCluster.numberOfNodes} Nodes</td>
|
2022-09-28 16:13:46 +02:00
|
|
|
</tr>
|
2023-04-12 18:00:28 +02:00
|
|
|
<tr class="py-2">
|
2022-09-28 16:13:46 +02:00
|
|
|
<th scope="col">Flop Rate (Any) <Icon name="info-circle" class="p-1" style="cursor: help;" title="Flops[Any] = (Flops[Double] x 2) + Flops[Single]"/></th>
|
2023-03-30 15:21:35 +02:00
|
|
|
<td style="min-width: 100px;"><div class="col"><Progress value={flopRate[subCluster.name]} max={subCluster.flopRateSimd.value * subCluster.numberOfNodes}/></div></td>
|
2023-06-16 12:44:34 +02:00
|
|
|
<td>
|
|
|
|
{scaleNumbers(flopRate[subCluster.name],
|
|
|
|
(subCluster.flopRateSimd.value * subCluster.numberOfNodes),
|
|
|
|
flopRateUnitPrefix[subCluster.name])
|
|
|
|
}{flopRateUnitBase[subCluster.name]} [Max]
|
|
|
|
</td>
|
2022-09-28 16:13:46 +02:00
|
|
|
</tr>
|
2023-04-12 18:00:28 +02:00
|
|
|
<tr class="py-2">
|
2022-09-28 16:13:46 +02:00
|
|
|
<th scope="col">MemBw Rate</th>
|
2023-03-30 15:21:35 +02:00
|
|
|
<td style="min-width: 100px;"><div class="col"><Progress value={memBwRate[subCluster.name]} max={subCluster.memoryBandwidth.value * subCluster.numberOfNodes}/></div></td>
|
2023-06-16 12:44:34 +02:00
|
|
|
<td>
|
|
|
|
{scaleNumbers(memBwRate[subCluster.name],
|
|
|
|
(subCluster.memoryBandwidth.value * subCluster.numberOfNodes),
|
|
|
|
memBwRateUnitPrefix[subCluster.name])
|
|
|
|
}{memBwRateUnitBase[subCluster.name]} [Max]
|
|
|
|
</td>
|
2022-09-28 16:13:46 +02:00
|
|
|
</tr>
|
|
|
|
</Table>
|
|
|
|
</CardBody>
|
|
|
|
</Card>
|
|
|
|
</Col>
|
|
|
|
<Col class="px-3">
|
|
|
|
<div bind:clientWidth={plotWidths[i]}>
|
|
|
|
{#key $mainQuery.data.nodeMetrics}
|
|
|
|
<Roofline
|
2022-10-07 15:01:14 +02:00
|
|
|
width={plotWidths[i] - 10} height={300} colorDots={true} showTime={false} cluster={subCluster}
|
2022-09-28 16:13:46 +02:00
|
|
|
data={transformPerNodeData($mainQuery.data.nodeMetrics.filter(data => data.subCluster == subCluster.name))} />
|
|
|
|
{/key}
|
|
|
|
</div>
|
2022-06-22 11:20:57 +02:00
|
|
|
</Col>
|
|
|
|
</Row>
|
|
|
|
{/each}
|
2022-09-28 16:13:46 +02:00
|
|
|
|
|
|
|
<hr style="margin-top: -1em;">
|
|
|
|
|
|
|
|
<!-- Usage Stats as Histograms -->
|
|
|
|
|
|
|
|
<Row cols={4}>
|
|
|
|
<Col class="p-2">
|
|
|
|
<div bind:clientWidth={colWidth1}>
|
|
|
|
<h4 class="mb-3 text-center">Top Users</h4>
|
|
|
|
{#key $mainQuery.data}
|
|
|
|
<Histogram
|
2023-03-30 15:21:35 +02:00
|
|
|
width={colWidth1 - 25}
|
2022-09-28 16:13:46 +02:00
|
|
|
data={$mainQuery.data.topUsers.sort((a, b) => b.count - a.count).map(({ count }, idx) => ({ count, value: idx }))}
|
2022-09-30 17:00:15 +02:00
|
|
|
label={(x) => x < $mainQuery.data.topUsers.length ? $mainQuery.data.topUsers[Math.floor(x)].name : '0'}
|
|
|
|
xlabel="User Name" ylabel="Number of Jobs" />
|
2022-09-28 16:13:46 +02:00
|
|
|
{/key}
|
|
|
|
</div>
|
|
|
|
</Col>
|
|
|
|
<Col class="px-4 py-2">
|
2022-06-22 11:20:57 +02:00
|
|
|
<Table>
|
2022-09-28 16:13:46 +02:00
|
|
|
<tr class="mb-2"><th>User Name</th><th>Number of Nodes</th></tr>
|
2022-06-22 11:20:57 +02:00
|
|
|
{#each $mainQuery.data.topUsers.sort((a, b) => b.count - a.count) as { name, count }}
|
|
|
|
<tr>
|
2023-06-12 12:12:15 +02:00
|
|
|
<th scope="col"><a href="/monitoring/user/{name}?cluster={cluster}&state=running">{name}</a></th>
|
2022-06-22 11:20:57 +02:00
|
|
|
<td>{count}</td>
|
|
|
|
</tr>
|
|
|
|
{/each}
|
|
|
|
</Table>
|
2022-09-28 16:13:46 +02:00
|
|
|
</Col>
|
|
|
|
<Col class="p-2">
|
|
|
|
<h4 class="mb-3 text-center">Top Projects</h4>
|
2022-06-22 11:20:57 +02:00
|
|
|
{#key $mainQuery.data}
|
|
|
|
<Histogram
|
2023-03-30 15:21:35 +02:00
|
|
|
width={colWidth1 - 25}
|
2022-06-22 11:20:57 +02:00
|
|
|
data={$mainQuery.data.topProjects.sort((a, b) => b.count - a.count).map(({ count }, idx) => ({ count, value: idx }))}
|
2022-09-30 17:00:15 +02:00
|
|
|
label={(x) => x < $mainQuery.data.topProjects.length ? $mainQuery.data.topProjects[Math.floor(x)].name : '0'}
|
|
|
|
xlabel="Project Code" ylabel="Number of Jobs" />
|
2022-06-22 11:20:57 +02:00
|
|
|
{/key}
|
2022-09-28 16:13:46 +02:00
|
|
|
</Col>
|
|
|
|
<Col class="px-4 py-2">
|
2022-06-22 11:20:57 +02:00
|
|
|
<Table>
|
2022-09-28 16:13:46 +02:00
|
|
|
<tr class="mb-2"><th>Project Code</th><th>Number of Nodes</th></tr>
|
2022-06-22 11:20:57 +02:00
|
|
|
{#each $mainQuery.data.topProjects.sort((a, b) => b.count - a.count) as { name, count }}
|
2023-04-12 18:00:28 +02:00
|
|
|
<tr>
|
2023-06-12 12:12:15 +02:00
|
|
|
<th scope="col"><a href="/monitoring/jobs/?cluster={cluster}&state=running&project={name}&projectMatch=eq">{name}</a></th>
|
2023-04-12 18:00:28 +02:00
|
|
|
<td>{count}</td>
|
|
|
|
</tr>
|
2022-06-22 11:20:57 +02:00
|
|
|
{/each}
|
|
|
|
</Table>
|
2022-09-28 16:13:46 +02:00
|
|
|
</Col>
|
2022-06-22 11:20:57 +02:00
|
|
|
</Row>
|
2022-09-28 16:13:46 +02:00
|
|
|
<Row cols={2} class="mt-3">
|
|
|
|
<Col class="p-2">
|
|
|
|
<div bind:clientWidth={colWidth2}>
|
|
|
|
<h4 class="mb-3 text-center">Duration Distribution</h4>
|
|
|
|
{#key $mainQuery.data.stats}
|
|
|
|
<Histogram
|
2023-03-30 15:21:35 +02:00
|
|
|
width={colWidth2 - 25}
|
2022-09-30 17:00:15 +02:00
|
|
|
data={$mainQuery.data.stats[0].histDuration}
|
2023-03-30 15:21:35 +02:00
|
|
|
xlabel="Current Runtimes [h]"
|
|
|
|
ylabel="Number of Jobs" />
|
2022-09-28 16:13:46 +02:00
|
|
|
{/key}
|
|
|
|
</div>
|
|
|
|
</Col>
|
|
|
|
<Col class="p-2">
|
|
|
|
<h4 class="mb-3 text-center">Number of Nodes Distribution</h4>
|
2022-06-22 11:20:57 +02:00
|
|
|
{#key $mainQuery.data.stats}
|
|
|
|
<Histogram
|
2023-03-30 15:21:35 +02:00
|
|
|
width={colWidth2 - 25}
|
2022-09-30 17:00:15 +02:00
|
|
|
data={$mainQuery.data.stats[0].histNumNodes}
|
2023-03-30 15:21:35 +02:00
|
|
|
xlabel="Allocated Nodes [#]"
|
|
|
|
ylabel="Number of Jobs" />
|
2022-06-22 11:20:57 +02:00
|
|
|
{/key}
|
2022-09-28 16:13:46 +02:00
|
|
|
</Col>
|
2022-06-22 11:20:57 +02:00
|
|
|
</Row>
|
|
|
|
{/if}
|