add metric family (#42)

* add metric family

the metric set designs do not necessarily map to an explicit
type, so I am adding family to allow for better categorization.

Signed-off-by: vsoch <[email protected]>

Author: vsoch

docs/_static/data/metrics.json

@@ -2,69 +2,79 @@
  {
   "name": "app-amg",
   "description": "parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids",
+  "family": "simulation",
   "type": "standalone",
   "image": "ghcr.io/converged-computing/metric-amg:latest",
   "url": "https://github.com/LLNL/AMG"
  },
  {
   "name": "app-kripke",
   "description": "parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids",
+  "family": "simulation",
   "type": "standalone",
   "image": "ghcr.io/converged-computing/metric-kripke:latest",
   "url": "https://github.com/LLNL/Kripke"
  },
  {
   "name": "app-lammps",
   "description": "LAMMPS molecular dynamic simulation",
+  "family": "simulation",
   "type": "standalone",
   "image": "ghcr.io/converged-computing/metric-lammps:latest",
   "url": "https://www.lammps.org/"
  },
  {
   "name": "app-pennant",
   "description": "Unstructured mesh hydrodynamics for advanced architectures ",
+  "family": "simulation",
   "type": "standalone",
   "image": "ghcr.io/converged-computing/metric-pennant:latest",
   "url": "https://github.com/LLNL/pennant"
  },
  {
   "name": "app-quicksilver",
   "description": "A proxy app for the Monte Carlo Transport Code",
+  "family": "simulation",
   "type": "standalone",
   "image": "ghcr.io/converged-computing/metric-quicksilver:latest",
   "url": "https://github.com/LLNL/Quicksilver"
  },
  {
   "name": "io-fio",
   "description": "Flexible IO Tester (FIO)",
+  "family": "storage",
   "type": "storage",
   "image": "ghcr.io/converged-computing/metric-fio:latest",
   "url": "https://fio.readthedocs.io/en/latest/fio_doc.html"
  },
  {
   "name": "io-sysstat",
   "description": "statistics for Linux tasks (processes) : I/O, CPU, memory, etc.",
+  "family": "storage",
   "type": "storage",
   "image": "ghcr.io/converged-computing/metric-sysstat:latest",
   "url": "https://github.com/sysstat/sysstat"
  },
  {
   "name": "network-netmark",
   "description": "point to point networking tool",
+  "family": "network",
   "type": "standalone",
   "image": "vanessa/netmark:latest",
   "url": ""
  },
  {
   "name": "network-osu-benchmark",
   "description": "point to point MPI benchmarks",
+  "family": "network",
   "type": "standalone",
   "image": "ghcr.io/converged-computing/metric-osu-benchmark:latest",
   "url": "https://mvapich.cse.ohio-state.edu/benchmarks/"
  },
  {
   "name": "perf-sysstat",
   "description": "statistics for Linux tasks (processes) : I/O, CPU, memory, etc.",
+  "family": "performance",
   "type": "application",
   "image": "ghcr.io/converged-computing/metric-sysstat:latest",
   "url": "https://github.com/sysstat/sysstat"

docs/_static/data/table.html

@@ -416,6 +416,7 @@
         <tr>
             <th>Name</th>
             <th>Type</th>
+            <th>Family</th>
             <th>Description</th>
             <th>Container</th>
         </tr>
@@ -440,6 +441,7 @@
           return "<a href='" + row['url'] + "' target='_blank'>" + data +"</a>";},
       },
       { data: "type"},
+      { data: "family"},
       { data: "description"},
       { data: "image",
         render: function ( data, type, row ) {
@@ -453,14 +455,18 @@
       }}
     ],
     'rowCallback': function(row, data, index){
-      if(data.type == 'storage'){
-        $(row).find('td:eq(1)').css('background-color', '#f5f580'); // yellow
+      // Distinguish family
+      if(data.family == 'storage'){
+        $(row).find('td:eq(2)').css('background-color', 'lavender');
       }
-      if(data.type == 'standalone'){
-        $(row).find('td:eq(1)').css('background-color', '#f79fb7'); // pinkish
+      if(data.family == 'performance'){
+        $(row).find('td:eq(2)').css('background-color', '#f79fb7');
       }
-      if(data.type == 'application'){
-        $(row).find('td:eq(1)').css('background-color', '#8af98a'); // lime green
+      if(data.family == 'network'){
+        $(row).find('td:eq(2)').css('background-color', 'orange');
+      }
+      if(data.family == 'simulation'){
+        $(row).find('td:eq(2)').css('background-color', 'skyblue');
       }
     }
   });

docs/getting_started/metrics.md

@@ -7,7 +7,11 @@ The following metrics are under development (or being planned).
  - [Application Metrics](https://converged-computing.github.io/metrics-operator/getting_started/metrics.html#application)
  - [Standalone Metrics](https://converged-computing.github.io/metrics-operator/getting_started/metrics.html#standalone)
 
-<iframe src="../_static/data/table.html" style="width:100%; height:700px;" frameBorder="0"></iframe>
+Each of the above is a metric design, which is primarily represented in the Metrics Operator code. However, within each design
+there are different families of metrics (e.g., storage, network, performance, simulation) shown in the table below as the "Family" column. 
+We likely will tweak and improve upon these categories.
+
+<iframe src="../_static/data/table.html" style="width:100%; height:800px;" frameBorder="0"></iframe>
 
 All metrics can be customized with the following variables
 
@@ -18,7 +22,8 @@ All metrics can be customized with the following variables
 
 ## Implemented Metrics
 
-Each metric has a link to the type, along with (optionally) examples.
+Each metric has a link to the type, along with (optionally) examples. These sections will better be organized by
+family once we decide on a more final set.
 
 ### Performance
 

hack/docs-gen/main.go

@@ -2,7 +2,6 @@ package main
 
 import (
 	"encoding/json"
-	"io/ioutil"
 	"log"
 	"os"
 	"sort"
@@ -23,6 +22,7 @@ var (
 type MetricOutput struct {
 	Name        string `json:"name"`
 	Description string `json:"description"`
+	Family      string `json:"family"`
 	Type        string `json:"type"`
 	Image       string `json:"image"`
 	Url         string `json:"url"`
@@ -38,6 +38,7 @@ func main() {
 		newRecord := MetricOutput{
 			Name:        metric.Name(),
 			Description: metric.Description(),
+			Family:      metric.Family(),
 			Type:        metric.Type(),
 			Image:       metric.Image(),
 			Url:         metric.Url(),
@@ -54,7 +55,7 @@ func main() {
 	if err != nil {
 		log.Fatalf("Could not marshall records %s\n", err.Error())
 	}
-	err = ioutil.WriteFile(filename, file, 0644)
+	err = os.WriteFile(filename, file, 0644)
 	if err != nil {
 		log.Fatalf("Could not write to file %s: %s\n", filename, err.Error())
 	}

pkg/jobs/application.go

@@ -0,0 +1,82 @@
+/*
+Copyright 2023 Lawrence Livermore National Security, LLC
+ (c.f. AUTHORS, NOTICE.LLNS, COPYING)
+
+SPDX-License-Identifier: MIT
+*/
+
+package jobs
+
+import (
+	api "github.com/converged-computing/metrics-operator/api/v1alpha1"
+	metrics "github.com/converged-computing/metrics-operator/pkg/metrics"
+	"k8s.io/apimachinery/pkg/util/intstr"
+	jobset "sigs.k8s.io/jobset/api/jobset/v1alpha2"
+)
+
+// These are common templates for application metrics
+
+// SingleApplication is a Metric base for a simple application metric
+// be accessible by other packages (and not conflict with function names)
+type SingleApplication struct {
+	Identifier    string
+	Rate          int32
+	Summary       string
+	Completions   int32
+	Container     string
+	Workdir       string
+	ResourceSpec  *api.ContainerResources
+	AttributeSpec *api.ContainerSpec
+}
+
+// Name returns the metric name
+func (m SingleApplication) Name() string {
+	return m.Identifier
+}
+
+// Description returns the metric description
+func (m SingleApplication) Description() string {
+	return m.Summary
+}
+
+// Default SingleApplication is generic performance family
+func (m SingleApplication) Family() string {
+	return metrics.PerformanceFamily
+}
+
+// Return container resources for the metric container
+func (m SingleApplication) Resources() *api.ContainerResources {
+	return m.ResourceSpec
+}
+func (m SingleApplication) Attributes() *api.ContainerSpec {
+	return m.AttributeSpec
+}
+
+// Validation
+func (m SingleApplication) Validate(spec *api.MetricSet) bool {
+	return true
+}
+
+// Container variables
+func (m SingleApplication) Image() string {
+	return m.Container
+}
+func (m SingleApplication) WorkingDir() string {
+	return m.Workdir
+}
+
+func (m SingleApplication) ReplicatedJobs(spec *api.MetricSet) ([]jobset.ReplicatedJob, error) {
+	return []jobset.ReplicatedJob{}, nil
+}
+
+func (m SingleApplication) ListOptions() map[string][]intstr.IntOrString {
+	return map[string][]intstr.IntOrString{}
+}
+
+func (m SingleApplication) SuccessJobs() []string {
+	return []string{}
+}
+
+func (m SingleApplication) Type() string {
+	return metrics.ApplicationMetric
+}

pkg/jobs/launcher.go

@@ -59,6 +59,11 @@ func (m LauncherWorker) Description() string {
 	return m.Summary
 }
 
+// Family returns a generic performance family
+func (m LauncherWorker) Family() string {
+	return metrics.PerformanceFamily
+}
+
 // Jobs required for success condition (n is the LauncherWorker run)
 func (m *LauncherWorker) SuccessJobs() []string {
 	m.ensureDefaultNames()

pkg/jobs/storage.go

@@ -35,6 +35,11 @@ func (m StorageGeneric) Name() string {
 	return m.Identifier
 }
 
+// Family returns the storage family
+func (m StorageGeneric) Family() string {
+	return metrics.StorageFamily
+}
+
 // Description returns the metric description
 func (m StorageGeneric) Description() string {
 	return m.Summary

pkg/metrics/app/amg.go

@@ -31,6 +31,11 @@ func (m AMG) Url() string {
 	return "https://github.com/LLNL/AMG"
 }
 
+// I think this is a simulation?
+func (m AMG) Family() string {
+	return metrics.SimulationFamily
+}
+
 // Set custom options / attributes for the metric
 func (m *AMG) SetOptions(metric *api.Metric) {
 	m.Rate = metric.Rate

pkg/metrics/app/kripke.go

@@ -29,6 +29,11 @@ func (m Kripke) Url() string {
 	return "https://github.com/LLNL/Kripke"
 }
 
+// I think this is a simulation?
+func (m Kripke) Family() string {
+	return metrics.SimulationFamily
+}
+
 // Set custom options / attributes for the metric
 func (m *Kripke) SetOptions(metric *api.Metric) {
 	m.Rate = metric.Rate

pkg/metrics/app/lammps.go

@@ -28,6 +28,11 @@ func (m Lammps) Url() string {
 	return "https://www.lammps.org/"
 }
 
+// I think this is a simulation?
+func (m Lammps) Family() string {
+	return metrics.SimulationFamily
+}
+
 // Set custom options / attributes for the metric
 func (m *Lammps) SetOptions(metric *api.Metric) {
 	m.Rate = metric.Rate