Simplify config handling and other fixes to orchestration

.gitignore

@@ -21,7 +21,6 @@ __debug_bin
 /hotstuff
 /plot
 
-
 # other
 *.in
 
@@ -35,3 +34,5 @@ measurements.json
 *.pdf
 
 twins.json
+internal/tree/benchmarkdata/*
+eventloop/benchmarkdata/*

.vscode/dict.txt

@@ -2,6 +2,7 @@ anypb
 Bano
 Baudet
 bbchain
+benchmarkdata
 BFT's
 Bitfield
 bitmask

.vscode/launch.json

@@ -12,7 +12,8 @@
 			"program": "${workspaceRoot}/cmd/hotstuff/main.go",
 			"cwd": "${workspaceRoot}",
 			"args": [
-				"run"
+				"run",
+				"--cue=scripts/local_config.cue"
 			]
 		},
 		{
@@ -55,4 +56,4 @@
 			"host": "127.0.0.1"
 		}
 	]
-}
+}

go.mod

@@ -11,7 +11,7 @@ require (
 	github.com/mitchellh/go-homedir v1.1.0
 	github.com/mroth/weightedrand v1.0.0
 	github.com/relab/gorums v0.7.1-0.20220818130557-8533cb369cd6
-	github.com/relab/iago v0.0.0-20240309193349-5be211223dec
+	github.com/relab/iago v0.0.0-20250112171319-2c4979b99287
 	github.com/relab/wrfs v0.0.0-20220416082020-a641cd350078
 	github.com/spf13/cobra v1.8.1
 	github.com/spf13/viper v1.19.0

go.sum

@@ -212,6 +212,8 @@ github.com/relab/gorums v0.7.1-0.20220818130557-8533cb369cd6 h1:azahqG2RhvhFvHiJ
 github.com/relab/gorums v0.7.1-0.20220818130557-8533cb369cd6/go.mod h1:dS1JU8uB1QgQie2bvRPeJWWmIFLPyl5IU50YfWpYVBE=
 github.com/relab/iago v0.0.0-20240309193349-5be211223dec h1:pF3HpDgnvSEm1kqShUAeYDH9mo+cWqo7/VKHsEU9ibc=
 github.com/relab/iago v0.0.0-20240309193349-5be211223dec/go.mod h1:3gDmi27sJB5VKFObac3BRuojM8bgBFeRwY2vo5jhbbI=
+github.com/relab/iago v0.0.0-20250112171319-2c4979b99287 h1:VqmwluYk/pXSqofd//0yeSey0BAdZG9uYwAG1naKuXo=
+github.com/relab/iago v0.0.0-20250112171319-2c4979b99287/go.mod h1:5aSZQxvFPFVU9nphdi+Q8hqN9N2XkW3/roQOlYhF/Vs=
 github.com/relab/wrfs v0.0.0-20220416082020-a641cd350078 h1:JN5qn8C/HZoyMAycX6z6O0SeX+09CV3w3GcVNW70OZA=
 github.com/relab/wrfs v0.0.0-20220416082020-a641cd350078/go.mod h1:8BTalsvE1BexSfZZFmZfmJKqfoXPzsP/ixpVOmg7Udk=
 github.com/rogpeppe/fastuuid v1.2.0/go.mod h1:jVj6XXZzXRy/MSR5jhDC/2q6DgLz+nrA6LYCDYWNEvQ=

internal/cli/flags.go

@@ -0,0 +1,80 @@
+package cli
+
+import (
+	"math"
+	"time"
+
+	"github.com/spf13/cobra"
+	"github.com/spf13/viper"
+)
+
+// runCmd represents the run command
+var runCmd = &cobra.Command{
+	Use:   "run",
+	Short: "Run an experiment.",
+	Long: `The run command runs an experiment locally or on remote workers.
+By default, a local experiment is run.
+To run experiments on remote machines, you must create a 'ssh_config' file (or use ~/.ssh/config).
+This should at the very least specify the identity file to use.
+It is also required that the host keys for all remote machines are present in a 'known_hosts' file.
+Then, you must use the '--ssh-config' parameter to specify the location of your 'ssh_config' file
+(or omit it to use ~/.ssh/config). Then, you must specify the list of remote machines to connect to
+using the '--host' parameter. This should be a comma separated list of hostnames or ip addresses.`,
+	Run: func(_ *cobra.Command, _ []string) {
+		runController()
+	},
+}
+
+func init() {
+	rootCmd.AddCommand(runCmd)
+
+	runCmd.Flags().Int("replicas", 4, "number of replicas to run")
+	runCmd.Flags().Int("clients", 1, "number of clients to run")
+	runCmd.Flags().Int("batch-size", 1, "number of commands to batch together in each block")
+	runCmd.Flags().Int("payload-size", 0, "size in bytes of the command payload")
+	runCmd.Flags().Int("max-concurrent", 4, "maximum number of concurrent commands per client")
+	runCmd.Flags().Duration("client-timeout", 500*time.Millisecond, "Client timeout.")
+	runCmd.Flags().Duration("duration", 10*time.Second, "duration of the experiment")
+	runCmd.Flags().Duration("connect-timeout", 5*time.Second, "duration of the initial connection timeout")
+	// need longer default timeout for the kauri
+	runCmd.Flags().Duration("view-timeout", 500*time.Millisecond, "duration of the first view")
+	runCmd.Flags().Duration("max-timeout", 0, "upper limit on view timeouts")
+	runCmd.Flags().Int("duration-samples", 1000, "number of previous views to consider when predicting view duration")
+	runCmd.Flags().Float32("timeout-multiplier", 1.2, "number to multiply the view duration by in case of a timeout")
+	runCmd.Flags().String("consensus", "chainedhotstuff", "name of the consensus implementation")
+	runCmd.Flags().String("crypto", "ecdsa", "name of the crypto implementation")
+	runCmd.Flags().String("leader-rotation", "round-robin", "name of the leader rotation algorithm")
+	runCmd.Flags().Int64("shared-seed", 0, "Shared random number generator seed")
+	runCmd.Flags().StringSlice("modules", nil, "Name additional modules to be loaded.")
+
+	runCmd.Flags().Bool("worker", false, "run a local worker")
+	runCmd.Flags().StringSlice("replica-hosts", nil, "the remote hosts to run replicas on via ssh")
+	runCmd.Flags().StringSlice("client-hosts", nil, "the remote hosts to run clients on via ssh")
+	runCmd.Flags().String("exe", "", "path to the executable to deploy and run on remote workers")
+	runCmd.Flags().String("ssh-config", "", "path to ssh_config file to resolve host aliases (defaults to ~/.ssh/config)")
+
+	runCmd.Flags().String("output", "", "the directory to save data and profiles to (disabled by default)")
+	runCmd.Flags().Bool("cpu-profile", false, "enable cpu profiling")
+	runCmd.Flags().Bool("mem-profile", false, "enable memory profiling")
+	runCmd.Flags().Bool("trace", false, "enable trace")
+	runCmd.Flags().Bool("fgprof-profile", false, "enable fgprof")
+
+	runCmd.Flags().StringSlice("metrics", []string{"client-latency", "throughput"}, "list of metrics to enable")
+	runCmd.Flags().Duration("measurement-interval", 0, "time interval between measurements")
+	runCmd.Flags().Float64("rate-limit", math.Inf(1), "rate limit for clients (in commands/second)")
+	runCmd.Flags().Float64("rate-step", 0, "rate limit step up for clients (in commands/second)")
+	runCmd.Flags().Duration("rate-step-interval", time.Hour, "how often the client rate limit should be increased")
+	runCmd.Flags().StringSlice("byzantine", nil, "byzantine strategies to use, as a comma separated list of 'name:count'")
+	// tree config parameter
+	runCmd.Flags().Int("bf", 2, "branch factor of the tree")
+	runCmd.Flags().IntSlice("tree-pos", []int{}, "tree positions of the replicas")
+	runCmd.Flags().Duration("tree-delta", 30*time.Millisecond, "waiting time for intermediate nodes in the tree")
+	runCmd.Flags().Bool("random-tree", false, "randomize tree positions, tree-pos is ignored if set")
+
+	runCmd.Flags().String("cue", "", "Cue-based host config")
+
+	err := viper.BindPFlags(runCmd.Flags())
+	if err != nil {
+		panic(err)
+	}
+}