Features for simulation with different controlplane architectures for arbitrary network topologies

CHANGELOG.md

@@ -2,6 +2,11 @@
 
 ## v0.5.1-dev
 
+- Simplify one of the switch protocols to avoid dependence on GraphMatching.jl which does not install well on non-linux systems. Do not rely on the default `SimpleSwitchDiscreteProt` for the time being.
+- Implement a network control protocol that is connection-oriented, centralized and non-distributed
+- Implement protocols: request generator and request tracker for simulation with the above control protocol in an asynchronous way.
+- Add `PhysicalGraph` struct for storing network metadata as the simulation evolves.
+- New tags: `EntanglementRequest`,`SwapRequest`, `DistributionRequest` and `RequestCompletion`
 - Add `classical_delay` and `quantum_delay` as keyword arguments to the `RegisterNet` constructor to set a default global network edge latency.
 
 ## v0.5.0 - 2024-10-16

examples/congestionchain/1_visualization.jl

@@ -44,7 +44,7 @@ scatter!(ax_fidZZ,ts,fidZZ,label="ZZ",color=(c2,0.1))
 
 display(fig)
 
-step_ts = range(0, 1000, step=0.1)
+step_ts = range(0.0, 1000.0, step=0.1)
 
 record(fig, "congestionchain.mp4", step_ts; framerate=50, visible=true) do t
     run(sim, t)

examples/controlplane/1a_cdd_interactive.jl

@@ -0,0 +1,46 @@
+include("setup.jl")
+
+succ_prob = Observable(0.001)
+for (;src, dst) in edges(net)
+    eprot = EntanglerProt(sim, net, src, dst; rounds=-1, randomize=true, success_prob=succ_prob[])
+    @process eprot()
+end
+
+local_busy_time = Observable(0.0)
+retry_lock_time = Observable(0.1)
+for node in 2:7
+    swapper = SwapperProt(sim, net, node; nodeL = <(node), nodeH = >(node), chooseL = argmin, chooseH = argmax, rounds=-1, local_busy_time=local_busy_time[],
+    retry_lock_time=retry_lock_time[])
+    @process swapper()
+end
+
+for v in vertices(net)
+    tracker = EntanglementTracker(sim, net, v)
+    @process tracker()
+end
+
+period_cons = Observable(0.1)
+consumer = EntanglementConsumer(sim, net, 1, 8; period=period_cons[])
+@process consumer()
+
+period_dec = Observable(0.1)
+for v in vertices(net)
+    cutoff = CutoffProt(sim, net, v; period=period_dec[])
+    @process cutoff()
+end
+params = [succ_prob, local_busy_time, retry_lock_time, period_cons, period_dec]
+sim, net, obs, entlog, entlogaxis, fid_axis, histaxis, num_epr_axis, fig = prepare_vis(consumer, params)
+
+step_ts = range(0.0, 1000.0, step=0.1)
+record(fig, "sim.mp4", step_ts; framerate=10, visible=true) do t
+    run(sim, t)
+    notify.((obs,entlog))
+    notify.(params)
+    ylims!(entlogaxis, (-1.04,1.04))
+    xlims!(entlogaxis, max(0,t-50), 1+t)
+    ylims!(fid_axis, (0, 1.04))
+    xlims!(fid_axis, max(0, t-50), 1+t)
+    autolimits!(histaxis)
+    ylims!(num_epr_axis, (0, 4))
+    xlims!(num_epr_axis, max(0, t-50), 1+t)
+end

examples/controlplane/1b_cdd_wglmakie.jl

@@ -0,0 +1,115 @@
+using WGLMakie
+WGLMakie.activate!()
+import Bonito
+using Markdown
+
+include("setup.jl")
+
+
+const custom_css = Bonito.DOM.style("ul {list-style: circle !important;}") # TODO remove after fix of bug in JSServe https://github.com/SimonDanisch/JSServe.jl/issues/178
+
+succ_prob = Observable(0.001)
+for (;src, dst) in edges(net)
+    eprot = EntanglerProt(sim, net, src, dst; rounds=-1, randomize=true, success_prob=succ_prob[])
+    @process eprot()
+end
+
+local_busy_time = Observable(0.0)
+retry_lock_time = Observable(0.1)
+for node in 2:7
+    swapper = SwapperProt(sim, net, node; nodeL = <(node), nodeH = >(node), chooseL = argmin, chooseH = argmax, rounds=-1, local_busy_time=local_busy_time[],
+    retry_lock_time=retry_lock_time[])
+    @process swapper()
+end
+
+for v in vertices(net)
+    tracker = EntanglementTracker(sim, net, v)
+    @process tracker()
+end
+
+period_cons = Observable(0.1)
+consumer = EntanglementConsumer(sim, net, 1, 8; period=period_cons[])
+@process consumer()
+
+period_dec = Observable(0.1)
+for v in vertices(net)
+    cutoff = CutoffProt(sim, net, v; period=period_dec[])
+    @process cutoff()
+end
+params = [succ_prob, local_busy_time, retry_lock_time, period_cons, period_dec]
+
+# All the calls that happen in the main event loop of the simulation,
+# encapsulated here so that we can conveniently pause the simulation from the WGLMakie app.
+function continue_singlerun!(sim, obs, entlog, params, entlogaxis, fid_axis, histaxis, num_epr_axis, running)
+    step_ts = range(0, 1000, step=0.1)
+    for t in step_ts
+        run(sim, t)
+        notify.((obs,entlog))
+        notify.(params) 
+        ylims!(entlogaxis, (-1.04,1.04))
+        xlims!(entlogaxis, max(0,t-50), 1+t)
+        ylims!(fid_axis, (0, 1.04))
+        xlims!(fid_axis, max(0, t-50), 1+t)
+        autolimits!(histaxis)
+        ylims!(num_epr_axis, (0, 4))
+        xlims!(num_epr_axis, max(0, t-50), 1+t)
+    end
+    running[] = nothing
+end
+
+#
+landing = Bonito.App() do
+
+    sim, net, obs, entlog, entlogaxis, fid_axis, histaxis, num_epr_axis, fig = prepare_vis(consumer, params)
+
+    running = Observable{Any}(false)
+    fig[5,1] = buttongrid = GridLayout(tellwidth = false)
+    buttongrid[1,1] = b = Makie.Button(fig, label = @lift(isnothing($running) ? "Done" : $running ? "Running..." : "Run once"), height=30, tellwidth=false)
+
+    on(b.clicks) do _
+        if !running[]
+            running[] = true
+        end
+    end
+    on(running) do r
+        if r
+            Threads.@spawn begin
+                continue_singlerun!(
+                    sim, obs, entlog, params, entlogaxis, fid_axis, histaxis, num_epr_axis, running)
+            end
+        end
+    end
+
+
+    content = md"""
+    Pick simulation settings and hit run (see below for technical details).
+
+    $(fig.scene)
+
+    # Connectionless, Distributed and Decentralized Control Plane for Entanglement Distribution
+
+    The above simulation visualizes entanglement distribution between Alice and Bob on an arbitrary network topology
+    given by the adjacency matrix of the graph. The control plane architecture used for this simulation is connectionless,
+    distributed and decentralized. The node representing Alice is the node on the top left and the bottom right is Bob.
+    The actual connectivity of the physical graph isn't fully captured by the visualization above as we use edges only to
+    show the virtual graph.
+
+   [See and modify the code for this simulation on github.](https://github.com/QuantumSavory/QuantumSavory.jl/tree/master/examples/controlplane/1b_cdd_wglmakie.jl)
+    """
+    return Bonito.DOM.div(Bonito.MarkdownCSS, Bonito.Styling, custom_css, content)
+end;
+
+#
+# Serve the Makie app
+
+isdefined(Main, :server) && close(server);
+port = parse(Int, get(ENV, "CDD_PORT", "8888"))
+interface = get(ENV, "CDD_IP", "127.0.0.1")
+proxy_url = get(ENV, "CDD_PROXY", "")
+server = Bonito.Server(interface, port; proxy_url);
+Bonito.HTTPServer.start(server)
+Bonito.route!(server, "/" => landing);
+
+##
+
+wait(server)

examples/controlplane/2a_cnc_interactive.jl

@@ -0,0 +1,40 @@
+include("setup.jl")
+
+controller = Controller(sim, net, 6, zeros(8,8))
+@process controller()
+
+req_gen = RequestGenerator(sim, net, 1, 8, 6)
+@process req_gen()
+
+consumer = EntanglementConsumer(sim, net, 1, 8)
+@process consumer()
+
+for node in 1:7
+    tracker = RequestTracker(sim, net, node)
+    @process tracker()
+end
+
+for v in 1:8
+    tracker = EntanglementTracker(sim, net, v)
+    @process tracker()
+end
+
+for v in 1:8
+    c_prot = CutoffProt(sim, net, v)
+    @process c_prot()
+end
+
+sim, net, obs, entlog, entlogaxis, fid_axis, histaxis, num_epr_axis, fig = prepare_vis(consumer)
+
+step_ts = range(0.0, 1000.0, step=0.1)
+record(fig, "sim.mp4", step_ts; framerate=10, visible=true) do t
+    run(sim, t)
+    notify.((obs,entlog))
+    ylims!(entlogaxis, (-1.04,1.04))
+    xlims!(entlogaxis, max(0,t-50), 1+t)
+    ylims!(fid_axis, (0, 1.04))
+    xlims!(fid_axis, max(0, t-50), 1+t)
+    autolimits!(histaxis)
+    ylims!(num_epr_axis, (0, 4))
+    xlims!(num_epr_axis, max(0, t-50), 1+t)
+end

examples/controlplane/2b_cnc_wglmakie.jl

@@ -0,0 +1,107 @@
+using WGLMakie
+WGLMakie.activate!()
+import Bonito
+using Markdown
+
+include("setup.jl")
+
+const custom_css = Bonito.DOM.style("ul {list-style: circle !important;}") # TODO remove after fix of bug in JSServe https://github.com/SimonDanisch/JSServe.jl/issues/178
+
+controller = Controller(sim, net, 6, zeros(8,8))
+@process controller()
+
+req_gen = RequestGenerator(sim, net, 1, 8, 6)
+@process req_gen()
+
+consumer = EntanglementConsumer(sim, net, 1, 8)
+@process consumer()
+
+for node in 1:7
+    tracker = RequestTracker(sim, net, node)
+    @process tracker()
+end
+
+for v in 1:8
+    tracker = EntanglementTracker(sim, net, v)
+    @process tracker()
+end
+
+for v in 1:8
+    c_prot = CutoffProt(sim, net, v)
+    @process c_prot()
+end
+
+# All the calls that happen in the main event loop of the simulation,
+# encapsulated here so that we can conveniently pause the simulation from the WGLMakie app.
+function continue_singlerun!(sim, obs, entlog, entlogaxis, fid_axis, histaxis, num_epr_axis, running)
+    step_ts = range(0, 1000, step=0.1)
+    for t in step_ts
+        run(sim, t)
+        notify.((obs,entlog)) 
+        ylims!(entlogaxis, (-1.04,1.04))
+        xlims!(entlogaxis, max(0,t-50), 1+t)
+        ylims!(fid_axis, (0, 1.04))
+        xlims!(fid_axis, max(0, t-50), 1+t)
+        autolimits!(histaxis)
+        ylims!(num_epr_axis, (0, 4))
+        xlims!(num_epr_axis, max(0, t-50), 1+t)
+    end
+    running[] = nothing
+end
+
+#
+landing = Bonito.App() do
+
+    sim, net, obs, entlog, entlogaxis, fid_axis, histaxis, num_epr_axis, fig = prepare_vis(consumer)
+
+    running = Observable{Any}(false)
+    fig[5,1] = buttongrid = GridLayout(tellwidth = false)
+    buttongrid[1,1] = b = Makie.Button(fig, label = @lift(isnothing($running) ? "Done" : $running ? "Running..." : "Run once"), height=30, tellwidth=false)
+
+    on(b.clicks) do _
+        if !running[]
+            running[] = true
+        end
+    end
+    on(running) do r
+        if r
+            Threads.@spawn begin
+                continue_singlerun!(
+                    sim, obs, entlog, entlogaxis, fid_axis, histaxis, num_epr_axis, running)
+            end
+        end
+    end
+
+
+    content = md"""
+    Pick simulation settings and hit run (see below for technical details).
+
+    $(fig.scene)
+
+    # Connection-Oriented, Non-Distributed and Centralized Control Plane for Entanglement Distribution
+
+    The above simulation visualizes entanglement distribution between Alice and Bob on an arbitrary network topology
+    given by the adjacency matrix of the graph. The control plane architecture used for this simulation is connection-oriented,
+    non-distributed and centralized. The node representing Alice is the node on the top left and the bottom right is Bob.
+    The actual connectivity of the physical graph isn't fully captured by the visualization above as we use edges only to
+    show the virtual graph.
+
+   [See and modify the code for this simulation on github.](https://github.com/QuantumSavory/QuantumSavory.jl/tree/master/examples/controlplane/2b_cnc_wglmakie.jl)
+    """
+    return Bonito.DOM.div(Bonito.MarkdownCSS, Bonito.Styling, custom_css, content)
+end;
+
+#
+# Serve the Makie app
+
+isdefined(Main, :server) && close(server);
+port = parse(Int, get(ENV, "CNC_PORT", "8888"))
+interface = get(ENV, "CNC_IP", "127.0.0.1")
+proxy_url = get(ENV, "CNC_PROXY", "")
+server = Bonito.Server(interface, port; proxy_url);
+Bonito.HTTPServer.start(server)
+Bonito.route!(server, "/" => landing);
+
+##
+
+wait(server)

examples/controlplane/3a_cl_interactive.jl

@@ -0,0 +1,45 @@
+include("setup.jl")
+
+for (;src, dst) in edges(net)
+    entangler = EntanglerProt(sim, net, src, dst; rounds=-1, randomize=true)
+    @process entangler()
+end
+
+controller = CLController(sim, net, 6)
+@process controller()
+
+req_gen = RequestGenerator(sim, net, 1, 8, 6)
+@process req_gen()
+
+consumer = EntanglementConsumer(sim, net, 1, 8)
+@process consumer()
+
+for node in 1:7
+    tracker = RequestTracker(sim, net, node)
+    @process tracker()
+end
+
+for v in 1:8
+    tracker = EntanglementTracker(sim, net, v)
+    @process tracker()
+end
+
+for v in 1:8
+    c_prot = CutoffProt(sim, net, v)
+    @process c_prot()
+end
+
+sim, net, obs, entlog, entlogaxis, fid_axis, histaxis, num_epr_axis, fig = prepare_vis(consumer)
+
+step_ts = range(0.0, 1000.0, step=0.1)
+record(fig, "sim.mp4", step_ts; framerate=10, visible=true) do t
+    run(sim, t)
+    notify.((obs,entlog))
+    ylims!(entlogaxis, (-1.04,1.04))
+    xlims!(entlogaxis, max(0,t-50), 1+t)
+    ylims!(fid_axis, (0, 1.04))
+    xlims!(fid_axis, max(0, t-50), 1+t)
+    autolimits!(histaxis)
+    ylims!(num_epr_axis, (0, 4))
+    xlims!(num_epr_axis, max(0, t-50), 1+t)
+end