Cytoscape.js GPU benchmarks

CPU	Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz
Cores	8 physical / 16 logical · 3.6 GHz base · 5 GHz max
Memory	62.6 GiB
GPU	CoffeeLake-S GT2 [UHD Graphics 630] i915 Ellesmere [Radeon RX 470/480/570/570X/580/580X/590] (8.0 GiB) amdgpu
OS	Fedora Linux 43 (Workstation Edition) · 6.19.14-200.fc43.x86_64 · x64
Node	v24.18.0 (v8 13.6.233.17-node.50)
Commit	c2d24c2e docs: the summary goes day by day, from the commit record
Machine id	5cf3f79c

Speedup overview v3 p50 ÷ gpu p50, log scale · right of the 1× line = gpu faster

sweep: edges() @ 2k

285107×

sweep: nodes() @ 2k

173123×

core: edges() @ 2k

93661×

core: nodes() @ 2k

50524×

sweep: elements() @ 2k

28736×

core: elements() @ 2k

20389×

core: graph scratch() set + get (v3 restyles the graph) @ 2k

16367×

core: removeScratch() @ 2k

8942×

compare: same() @ 2k

8155×

core: graph data() set + get (v3 restyles the graph) @ 2k

8137×

trav: predecessors() @ 2k

723×

trav: successors() @ 2k

719×

mut-bulk: hide + show @ 2k

268×

flags: show + hide + visible @ 2k

227×

parent drag (shift subtree + bb settle) @ 2k

185×

compound: commonAncestors @ 2k

124×

degree: nodes.totalDegree() @ 2k

112×

mut-bulk: shift @ 2k

111×

viewport: reset() @ 2k

105×

degree: nodes.maxDegree() @ 2k

96×

94×

flags: panify + unpanify + pannable @ 2k

89×

flags: selectify + unselectify + selectable @ 2k

88×

degree: node.degree() @ 2k

88×

flags: lock + unlock + locked @ 2k

88×

flags: grabify + ungrabify + grabbable @ 2k

87×

core: options() @ 2k

87×

degree: indegree + outdegree (2 calls) @ 2k

77×

reparent round-trip @ 2k

76×

collection: same + anySame + contains (3 calls) @ 2k

75×

algo: pageRank (20 iters) @ 500

70×

algo: closenessCentrality (one root) @ 2k

70×

algo: closenessCentrality (one root) @ 500

69×

mut-bulk: positions(obj) @ 2k

65×

dimensions: shift + silentShift (round-trip) @ 2k

61×

60×

algo: dfs (whole graph) @ 2k

59×

viewport: viewport({ zoom, pan }) @ 2k

59×

sweep: nodes(":selected") vs nodes({selected}) @ 2k

59×

algo: bfs (whole graph) @ 2k

58×

algo: bfs (whole graph) @ 500

57×

algo: dfs (whole graph) @ 500

56×

algo: dijkstra + pathTo @ 2k

56×

mut-bulk: lock + unlock @ 2k

54×

algo: closenessCentrality (weighted, one root) @ 500

53×

child drag (move + parent re-derive) @ 2k

51×

dimensions: positions( fn ) over 100 @ 2k

50×

algo: dijkstra + pathTo @ 500

49×

degree: minDegree + maxDegree (2 calls) @ 2k

44×

collection: component() @ 2k

43×

traverse: components() @ 2k

41×

algo: degreeCentralityNormalized @ 500

37×

algo: degreeCentralityNormalized @ 2k

37×

36×

sweep: selected nodes ($("node:selected") vs filter({group, selected})) @ 2k

35×

algo: degreeCentrality (one root) @ 500

35×

query: structural, parents (:parent vs { parent: true }) @ 2k

34×

algo: degreeCentrality (one root) @ 2k

34×

algo: tarjan SCC @ 500

34×

algo: tarjan SCC @ 2k

33×

bounds: getFitViewport() (compute without committing) @ 2k

33×

bounds: cy.fit() (whole-graph scan) @ 2k

33×

query: structural, orphans (:orphan vs { child: false }) @ 2k

32×

collection: silentPositions( fn ) over 100 @ 2k

30×

collection: difference @ 2k

29×

collection: intersection @ 2k

29×

algo: kruskal @ 2k

28×

emit: position write, no listeners @ 2k

28×

28×

algo: hopcroft-tarjan biconnected @ 500

26×

scn: select-all + fit (listeners on) @ 2k

26×

set: intersection() @ 2k

26×

load: init from the definition form (construct + dispose) @ 2k

26×

set: difference() @ 2k

26×

degree: minIndegree + maxOutdegree (2 calls) @ 2k

26×

degree: maxIndegree + minOutdegree (2 calls) @ 2k

25×

algo: kruskal @ 500

25×

algo: hopcroft-tarjan biconnected @ 2k

24×

trav: band roots() @ 2k

24×

core: id lookup ($("#id") vs $id()) @ 2k

23×

mut-bulk: positions(fn) @ 2k

23×

mut: position set @ 2k

23×

data: bulk write, object column (plain-array fallback) @ 2k

21×

algo: bellmanFord @ 500

20×

data: bulk write, numeric column @ 2k

20×

algo: bellmanFord @ 2k

20×

19×

core: add + remove @ 2k

18×

data: bulk write, one new dictionary entry per pass @ 2k

18×

viewport: center( eles ) @ 2k

18×

data: bulk write, dictionary string column (4 values) @ 2k

16×

scn: drag (select band of 100 + 8-step shift, position listener) @ 2k

16×

dimensions: silentPosition @ 2k

16×

query: membership ([bucket in ...] vs { in }) @ 2k

15×

collection: absoluteComplement @ 2k

15×

flags: active/activate/unactivate (3 calls) @ 2k

15×

flags: select + unselect + selected (band) @ 2k

15×

viewport: minZoom/maxZoom get (2 calls) @ 2k

15×

mut-bulk: data set @ 2k

14×

query: data equality ([bucket = 3] vs { data: { bucket: 3 } }) @ 2k

14×

core: selected ($(":selected") vs filter({selected})) @ 2k

14×

core: forceRender() (headless no-op) @ 2k

14×

style: renderedStyle (one prop) @ 2k

13×

query: structural, childless (:childless vs { parent: false }) @ 2k

13×

traversal: roots + leaves (2 calls) @ 2k

13×

sweep: selected ($(":selected") vs filter({selected})) @ 2k

13×

dimensions: padding + paddedWidth + paddedHeight (3 calls) @ 2k

13×

algo: betweennessCentrality (unweighted) @ 500

12×

query: two keys AND-ed @ 2k

12×

12×

collection: toArray() @ 2k

12×

algo: fuzzyCMeans @ 500

11×

dimensions: relativePosition get @ 2k

11×

algo: betweennessCentrality (unweighted) @ 2k

11×

query: data comparison ([weight > 3] vs { gt }) @ 2k

11×

11×

10×

sweep: all nodes ($("node") vs filter({group})) @ 2k

9.5×

collection: removeScratch() @ 2k

9.0×

algo: betweennessCentrality (weighted) @ 500

8.8×

algo: kMedoids @ 500

8.7×

core: removeAllListeners() @ 2k

8.5×

emit: 1 core listener @ 2k

8.5×

data: removeData (whole collection) @ 2k

8.4×

algo: hierarchicalClustering (threshold) @ 500

8.3×

core: batchData() @ 2k

8.0×

emit: delegated (v3 selector vs v4 predicate) @ 2k

8.0×

scn: explore (expand 2-hop + select + fit) @ 2k

8.0×

mut-bulk: select + unselect @ 2k

7.8×

element data: scratch get + set @ 2k

7.6×

emit: 1 element (ref-qualified) listener @ 2k

7.6×

core: collection() @ 2k

7.2×

core: collection() @ 2k

7.2×

compound: orphans + nonorphans (2 calls) @ 2k

6.9×

scn: edit (remove + re-add 256 nodes + cascade, add/remove listeners) @ 2k

6.8×

mut-bulk: remove + re-add (256 nodes + cascade) @ 2k

6.4×

core: batch( fn ) (empty) @ 2k

6.3×

style: whole-object style() @ 2k

6.2×

emit: 10 core listeners @ 2k

6.1×

query: predicate function (both sides) @ 2k

6.0×

collection: union @ 2k

6.0×

core: startBatch + endBatch + batching @ 2k

5.9×

bounds: whole-graph boundingBox() @ 2k

5.9×

style: bypass set + removeStyle round-trip @ 2k

5.8×

style: sheet swap (compile + applyAll) @ 2k

5.8×

traverse: outgoers() @ 2k

5.4×

traversal: allAreNeighbors @ 2k

5.3×

traverse: incomers() @ 2k

5.3×

algo: hierholzer (directed) @ 500

5.2×

set: union() @ 2k

5.1×

algo: affinityPropagation @ 500

5.1×

trav: band neighborhood() @ 2k

5.0×

trav: outgoers() @ 2k

4.9×

core: animated() @ 2k

4.9×

trav: incomers() @ 2k

4.9×

core: nodes ($("node") vs filter({group})) @ 2k

4.9×

compare: contains() @ 2k

4.9×

compound: children + descendants + siblings (3 calls) @ 2k

4.8×

viewport: zoom() get @ 2k

4.7×

algo: kMeans (fixed centroids) @ 500

4.6×

collection: length + empty + nonempty (3 calls) @ 2k

4.6×

style: bypass set (ele.style name,value) @ 2k

4.1×

mut: select + unselect @ 2k

4.0×

trav: neighborhood() @ 2k

3.9×

traverse: neighborhood() @ 2k

3.8×

collection: symmetricDifference @ 2k

3.7×

compound: parent + parents + ancestors (3 calls) @ 2k

3.7×

collection: cy() @ 2k

3.6×

collection: layoutDimensions() @ 2k

3.6×

algo: aStar @ 500

3.5×

traversal: parallelEdges + codirectedEdges (2 calls) @ 2k

3.5×

algo: aStar @ 2k

3.4×

traversal: sources + targets (2 calls) @ 2k

3.4×

collection: diff @ 2k

3.3×

traversal: connectedNodes @ 2k

3.3×

traversal: isLoop + isSimple (2 calls) @ 2k

3.1×

flags: takesUpSpace + interactive + transparent (3 calls) @ 2k

3.1×

collection: byGroup @ 2k

3.0×

anim: start + stop, 512-node collection @ 2k

3.0×

dimensions: rendered dims (4 calls) @ 2k

3.0×

anim: viewport animate + stop @ 2k

3.0×

traversal: openNeighborhood + closedNeighborhood (2 calls) @ 2k

2.9×

viewport: width() + height() + size() (3 calls) @ 2k

2.9×

collection: animated() @ 2k

2.9×

style: numericStyle @ 2k

2.9×

iter: map() @ 2k

2.9×

algo: hierholzer (directed) @ 2k

2.9×

trav: band connectedNodes() @ 2k

2.8×

core: window() @ 2k

2.8×

load: cy.json() @ 2k

2.8×

collection: delayAnimation() @ 2k

2.8×

traverse: connectedEdges() @ 2k

2.7×

mut: data set @ 2k

2.7×

dimensions: width + height + outerWidth + outerHeight (4 calls) @ 2k

2.7×

core: container() @ 2k

2.7×

trav: closedNeighborhood() @ 2k

2.6×

trav: connectedEdges() @ 2k

2.5×

collection: is (none match: the full walk) @ 2k

2.5×

core: onRender + offRender @ 2k

2.4×

dimensions: renderedPosition get @ 2k

2.4×

core: resize() / invalidateSize() @ 2k

2.4×

flags: effectiveOpacity @ 2k

2.3×

collection: eq/first/last (3 calls) @ 2k

2.3×

core: zoom set @ 2k

2.3×

core: destroyed() @ 2k

2.2×

emit: on() + off() round-trip @ 2k

2.2×

style: read one prop (width) @ 2k

2.2×

data: read the whole object @ 2k

2.2×

core: isReady() @ 2k

2.2×

collection: slice() @ 2k

2.2×

element data: json() one element @ 2k

2.1×

collection: min + max ( fn ) (2 calls) @ 2k

2.0×

core: selectionType get @ 2k

1.99×

element data: jsons() over 100 @ 2k

1.99×

flags: grabbed + hidden + inactive (3 calls) @ 2k

1.99×

style: first apply on add (256-node band) @ 2k

1.94×

dimensions: renderedBoundingBox @ 2k

1.93×

algo: floydWarshall @ 500

1.93×

viewport: zoom( n ) set @ 2k

1.90×

core: gating setters round-trip (autolock/autoungrabify/autounselectify) @ 2k

1.90×

json: element @ 2k

1.88×

collection: indexOfId() @ 2k

1.79×

scn: refresh (bulk data write + mapped labels + filter + fit, data listener) @ 2k

1.79×

viewport: pan( p ) set @ 2k

1.79×

algo: closenessCentralityNormalized @ 500

1.78×

iter: filter(fn) @ 2k

1.78×

collection: element accessors (id/group/isNode/isEdge, 4 calls) @ 2k

1.76×

style: read one prop (color) @ 2k

1.73×

style( 'text-max-width' ) (was #74 in the switch) @ 2k

1.73×

style( 'text-wrap' ) (was #73 in the switch) @ 2k

1.70×

sweep: filter(fn) @ 2k

1.69×

load: add a 256-node band (add + remove) @ 2k

1.66×

viewport: panBy() @ 2k

1.64×

position: get @ 2k

1.64×

viewport: renderedExtent() @ 2k

1.63×

collection: some + every (2 calls) @ 2k

1.62×

collection: sort( fn ) @ 2k

1.62×

bounds: one node boundingBox() @ 2k

1.61×

anim: delay() (no channels touched) @ 2k

1.59×

compound: isParent/isChild/isChildless/isOrphan (4 calls) @ 2k

1.56×

collection: allAre (all match: the full walk) @ 2k

1.55×

viewport: extent() @ 2k

1.50×

collection: indexOf @ 2k

1.50×

trav: band sources() @ 2k

1.48×

trav: band connectedEdges() @ 2k

1.46×

data: read one numeric key @ 2k

1.41×

collection: reduce( fn ) @ 2k

1.40×

anim: start + stop, one element @ 2k

1.37×

core: mutableElements() (v4 materializes; v3 is O(1)) @ 2k

1.36×

traversal: source() + target() (2 calls) @ 2k

1.35×

data: read one string key (dictionary decode) @ 2k

1.35×

core: pan() get (x32) @ 2k

1.31×

style( 'taxi-radius' ) (was #142 in the switch) @ 2k

1.29×

traversal: edgesTo @ 2k

1.28×

traversal: edgesWith @ 2k

1.27×

style( 'target-distance-from-node' ) (was #150 in the switch) @ 2k

1.27×

core: instanceString + headless + styleEnabled (3 calls) @ 2k

1.27×

core: hasElementWithId + hasCompoundNodes (2 calls) @ 2k

1.19×

core: gating getters (5 calls) @ 2k

1.18×

collection: removed() + inside() (2 calls) @ 2k

1.17×

style( 'border-width' ) (was #6 in the switch) @ 2k

1.15×

style( 'background-color' ) (was #4 in the switch) @ 2k

1.12×

core: extent() @ 2k

1.12×

algo: markovClustering @ 500

1.11×

viewport: pan() get @ 2k

1.09×

iter: forEach() @ 2k

1.04×

algo: kargerStein @ 500

1.04×

data: get @ 2k

1.03×

core: filter(fn) @ 2k

0.99×

core: getElementById() @ 2k

0.98×

0.1×1×10×100×1000×10000×100000×1000000×

Scaling speedup vs graph size

algorithms

group	× @ N=500	× @ N=2k
algo: bfs (whole graph)	57×	58×
algo: dfs (whole graph)	56×	59×
algo: dijkstra + pathTo	49×	56×
algo: aStar	3.5×	3.4×
algo: bellmanFord	20×	20×
algo: kruskal	25×	28×
algo: tarjan SCC	34×	33×
algo: hopcroft-tarjan biconnected	26×	24×
algo: hierholzer (directed)	5.2×	2.9×
algo: betweennessCentrality (unweighted)	12×	11×
algo: degreeCentralityNormalized	37×	37×
algo: closenessCentrality (one root)	69×	70×
algo: degreeCentrality (one root)	35×	34×
algo: pageRank (20 iters)	70×	–
algo: closenessCentrality (weighted, one root)	53×	–
algo: floydWarshall	1.93×	–
algo: closenessCentralityNormalized	1.78×	–
algo: markovClustering	1.11×	–
algo: hierarchicalClustering (threshold)	8.3×	–
algo: kMeans (fixed centroids)	4.6×	–
algo: kMedoids	8.7×	–
algo: fuzzyCMeans	11×	–
algo: affinityPropagation	5.1×	–
algo: kargerStein	1.04×	–
algo: betweennessCentrality (weighted)	8.8×	–

core+collection N=2,000 · 2.9 minv4 (gpu) v3

core: getElementById()

0.98×

core: id lookup ($("#id") vs $id())

23×

core: nodes()

50524×

core: edges()

93661×

core: elements()

20389×

core: nodes ($("node") vs filter({group}))

4.9×

core: selected ($(":selected") vs filter({selected}))

14×

core: filter(fn)

0.99×

core: collection()

7.2×

core: pan() get (x32)

1.31×

core: extent()

1.12×

core: add + remove

18×

core: zoom set

2.3×

traverse: neighborhood()

3.8×

traverse: connectedEdges()

2.7×

traverse: outgoers()

5.4×

traverse: incomers()

5.3×

traverse: components()

41×

degree: node.degree()

88×

degree: nodes.totalDegree()

112×

degree: nodes.maxDegree()

96×

data: get

1.03×

position: get

1.64×

json: element

1.88×

set: union()

5.1×

set: intersection()

26×

set: difference()

26×

iter: map()

2.9×

iter: forEach()

1.04×

iter: filter(fn)

1.78×

compare: same()

8155×

compare: contains()

4.9×

mut: data set

2.7×

mut: position set

23×

mut: select + unselect

4.0×

1 ns10 ns100 ns1 µs10 µs100 µs1 ms10 ms100 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
core: getElementById()	v3	89.2 ns	91.8 ns	131 ns	85.2 ns	1868
core: getElementById()	gpu	91.2 ns	94.2 ns	128 ns	87.2 ns	1817
core: id lookup ($("#id") vs $id())	v3	1.95 µs	2.30 µs	3.69 µs	1.82 µs	69
core: id lookup ($("#id") vs $id())	gpu	84.6 ns	87.7 ns	159 ns	82.1 ns	1950
core: nodes()	v3	412 µs	424 µs	569 µs	406 µs	1647
core: nodes()	gpu	8.15 ns	8.47 ns	18.8 ns	7.88 ns	20125
core: edges()	v3	725 µs	740 µs	945 µs	715 µs	945
core: edges()	gpu	7.74 ns	8.22 ns	15.1 ns	7.67 ns	20650
core: elements()	v3	142 µs	148 µs	252 µs	140 µs	4697
core: elements()	gpu	6.96 ns	7.30 ns	18.3 ns	6.63 ns	23282
core: nodes ($("node") vs filter({group}))	v3	548 µs	581 µs	1.22 ms	533 µs	1201
core: nodes ($("node") vs filter({group}))	gpu	113 µs	119 µs	276 µs	82.8 µs	5773
core: selected ($(":selected") vs filter({selected}))	v3	346 µs	359 µs	630 µs	338 µs	1937
core: selected ($(":selected") vs filter({selected}))	gpu	24.9 µs	25.5 µs	26.6 µs	24.5 µs	12
core: filter(fn)	v3	414 µs	427 µs	693 µs	399 µs	1637
core: filter(fn)	gpu	418 µs	452 µs	956 µs	410 µs	1533
core: collection()	v3	463 ns	570 ns	1.68 µs	447 ns	293
core: collection()	gpu	64.4 ns	68.5 ns	116 ns	61.9 ns	2431
core: pan() get (x32)	v3	44.5 ns	45.4 ns	55.9 ns	43.5 ns	3654
core: pan() get (x32)	gpu	33.9 ns	35.7 ns	50.6 ns	32.9 ns	4793
core: extent()	v3	22.9 ns	39.5 ns	142 ns	21.9 ns	4214
core: extent()	gpu	20.5 ns	28.1 ns	108 ns	19.1 ns	5917
core: add + remove	v3	121 µs	124 µs	196 µs	110 µs	5635
core: add + remove	gpu	6.77 µs	7.77 µs	18.8 µs	6.38 µs	80570
core: zoom set	v3	1.33 µs	1.34 µs	1.47 µs	1.20 µs	122
core: zoom set	gpu	583 ns	598 ns	774 ns	572 ns	272
traverse: neighborhood()	v3	3.70 µs	3.85 µs	4.47 µs	3.43 µs	40
traverse: neighborhood()	gpu	976 ns	1.11 µs	2.01 µs	938 ns	146
traverse: connectedEdges()	v3	1.05 µs	1.17 µs	2.04 µs	1.01 µs	141
traverse: connectedEdges()	gpu	385 ns	423 ns	1.36 µs	376 ns	395
traverse: outgoers()	v3	2.44 µs	2.61 µs	3.38 µs	2.32 µs	61
traverse: outgoers()	gpu	454 ns	538 ns	2.03 µs	442 ns	300
traverse: incomers()	v3	2.42 µs	2.61 µs	3.41 µs	2.35 µs	61
traverse: incomers()	gpu	460 ns	571 ns	2.03 µs	449 ns	283
traverse: components()	v3	80.2 ms	82.2 ms	90.8 ms	75.6 ms	12
traverse: components()	gpu	1.93 ms	2.07 ms	3.55 ms	1.88 ms	319
degree: node.degree()	v3	3.66 µs	3.78 µs	4.61 µs	3.27 µs	41
degree: node.degree()	gpu	41.5 ns	44.0 ns	130 ns	39.6 ns	3881
degree: nodes.totalDegree()	v3	7.27 ms	8.09 ms	12.0 ms	6.88 ms	82
degree: nodes.totalDegree()	gpu	64.7 µs	67.3 µs	120 µs	60.1 µs	10399
degree: nodes.maxDegree()	v3	7.25 ms	8.07 ms	12.8 ms	6.79 ms	83
degree: nodes.maxDegree()	gpu	75.2 µs	77.8 µs	138 µs	69.8 µs	9006
data: get	v3	30.0 ns	31.6 ns	60.1 ns	29.0 ns	5418
data: get	gpu	29.2 ns	30.9 ns	61.1 ns	28.5 ns	5524
position: get	v3	35.8 ns	37.1 ns	56.9 ns	34.4 ns	4610
position: get	gpu	21.8 ns	25.6 ns	54.1 ns	21.0 ns	6541
json: element	v3	300 ns	393 ns	3.52 µs	292 ns	423
json: element	gpu	160 ns	166 ns	266 ns	156 ns	982
set: union()	v3	93.3 µs	106 µs	171 µs	91.1 µs	6456
set: union()	gpu	18.1 µs	18.8 µs	18.7 µs	17.6 µs	12
set: intersection()	v3	92.6 µs	106 µs	155 µs	88.4 µs	6517
set: intersection()	gpu	3.56 µs	3.61 µs	3.93 µs	3.50 µs	42
set: difference()	v3	91.1 µs	105 µs	195 µs	88.4 µs	6576
set: difference()	gpu	3.54 µs	3.58 µs	3.98 µs	3.45 µs	43
iter: map()	v3	29.6 µs	31.6 µs	57.4 µs	28.5 µs	21839
iter: map()	gpu	10.3 µs	10.3 µs	10.7 µs	10.2 µs	12
iter: forEach()	v3	3.48 µs	3.73 µs	3.83 µs	3.43 µs	36
iter: forEach()	gpu	3.35 µs	3.40 µs	3.89 µs	3.28 µs	44
iter: filter(fn)	v3	245 µs	254 µs	346 µs	239 µs	2756
iter: filter(fn)	gpu	138 µs	143 µs	204 µs	135 µs	4885
compare: same()	v3	119 µs	124 µs	219 µs	114 µs	5678
compare: same()	gpu	14.6 ns	15.6 ns	30.8 ns	14.2 ns	10943
compare: contains()	v3	6.12 µs	6.19 µs	6.44 µs	6.04 µs	24
compare: contains()	gpu	1.26 µs	1.27 µs	1.37 µs	1.23 µs	128
mut: data set	v3	618 ns	645 ns	1.23 µs	601 ns	260
mut: data set	gpu	230 ns	241 ns	373 ns	229 ns	706
mut: position set	v3	1.18 µs	1.19 µs	1.27 µs	1.15 µs	136
mut: position set	gpu	52.1 ns	55.1 ns	100 ns	50.1 ns	3083
mut: select + unselect	v3	2.43 µs	2.61 µs	3.25 µs	2.34 µs	61
mut: select + unselect	gpu	610 ns	617 ns	672 ns	597 ns	263

materializers N=2,000 · 49.8 sv4 (gpu) v3

sweep: nodes()

173123×

sweep: edges()

285107×

sweep: elements()

28736×

sweep: all nodes ($("node") vs filter({group}))

9.5×

sweep: selected ($(":selected") vs filter({selected}))

13×

sweep: selected nodes ($("node:selected") vs filter({group, selected}))

35×

sweep: nodes(":selected") vs nodes({selected})

59×

sweep: filter(fn)

1.69×

1 ns10 ns100 ns1 µs10 µs100 µs1 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
sweep: nodes()	v3	417 µs	428 µs	568 µs	407 µs	1637
sweep: nodes()	gpu	2.41 ns	2.51 ns	5.29 ns	2.40 ns	65852
sweep: edges()	v3	717 µs	731 µs	927 µs	696 µs	956
sweep: edges()	gpu	2.51 ns	2.63 ns	4.89 ns	2.46 ns	62891
sweep: elements()	v3	139 µs	148 µs	235 µs	136 µs	4709
sweep: elements()	gpu	4.85 ns	5.19 ns	12.1 ns	4.64 ns	32562
sweep: all nodes ($("node") vs filter({group}))	v3	545 µs	576 µs	866 µs	527 µs	1213
sweep: all nodes ($("node") vs filter({group}))	gpu	57.5 µs	72.3 µs	176 µs	53.9 µs	9578
sweep: selected ($(":selected") vs filter({selected}))	v3	328 µs	338 µs	525 µs	322 µs	2070
sweep: selected ($(":selected") vs filter({selected}))	gpu	25.8 µs	26.4 µs	27.8 µs	25.6 µs	12
sweep: selected nodes ($("node:selected") vs filter({group, selected}))	v3	318 µs	328 µs	556 µs	313 µs	2134
sweep: selected nodes ($("node:selected") vs filter({group, selected}))	gpu	9.03 µs	9.11 µs	9.69 µs	8.86 µs	15
sweep: nodes(":selected") vs nodes({selected})	v3	529 µs	560 µs	1.12 ms	504 µs	1250
sweep: nodes(":selected") vs nodes({selected})	gpu	9.03 µs	9.06 µs	9.10 µs	8.97 µs	15
sweep: filter(fn)	v3	409 µs	430 µs	739 µs	402 µs	1628
sweep: filter(fn)	gpu	242 µs	260 µs	566 µs	237 µs	2680

mutators N=2,000 · 56.4 sv4 (gpu) v3

mut-bulk: select + unselect

7.8×

mut-bulk: hide + show

268×

mut-bulk: lock + unlock

54×

mut-bulk: positions(obj)

65×

mut-bulk: positions(fn)

23×

mut-bulk: shift

111×

mut-bulk: data set

14×

mut-bulk: remove + re-add (256 nodes + cascade)

6.4×

10 µs100 µs1 ms10 ms100 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
mut-bulk: select + unselect	v3	2.00 ms	2.06 ms	3.19 ms	1.95 ms	319
mut-bulk: select + unselect	gpu	255 µs	265 µs	400 µs	249 µs	2649
mut-bulk: hide + show	v3	57.7 ms	63.2 ms	72.2 ms	56.0 ms	12
mut-bulk: hide + show	gpu	216 µs	222 µs	396 µs	206 µs	3067
mut-bulk: lock + unlock	v3	2.09 ms	2.15 ms	4.13 ms	2.02 ms	320
mut-bulk: lock + unlock	gpu	38.7 µs	39.1 µs	40.5 µs	38.1 µs	12
mut-bulk: positions(obj)	v3	2.39 ms	2.45 ms	3.88 ms	2.19 ms	279
mut-bulk: positions(obj)	gpu	37.0 µs	38.6 µs	64.6 µs	35.7 µs	17865
mut-bulk: positions(fn)	v3	3.41 ms	3.58 ms	5.60 ms	3.23 ms	191
mut-bulk: positions(fn)	gpu	148 µs	153 µs	267 µs	143 µs	4469
mut-bulk: shift	v3	3.58 ms	3.75 ms	5.16 ms	3.49 ms	180
mut-bulk: shift	gpu	32.3 µs	33.7 µs	59.7 µs	31.2 µs	20424
mut-bulk: data set	v3	907 µs	930 µs	1.39 ms	880 µs	732
mut-bulk: data set	gpu	63.5 µs	65.1 µs	109 µs	61.7 µs	10440
mut-bulk: remove + re-add (256 nodes + cascade)	v3	41.4 ms	44.9 ms	54.5 ms	40.3 ms	12
mut-bulk: remove + re-add (256 nodes + cascade)	gpu	6.49 ms	6.64 ms	8.56 ms	5.73 ms	99

scenarios N=2,000 · 37.8 sv4 (gpu) v3

scn: explore (expand 2-hop + select + fit)

8.0×

scn: select-all + fit (listeners on)

26×

scn: drag (select band of 100 + 8-step shift, position listener)

16×

scn: edit (remove + re-add 256 nodes + cascade, add/remove listeners)

6.8×

scn: refresh (bulk data write + mapped labels + filter + fit, data listener)

1.79×

10 µs100 µs1 ms10 ms100 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
scn: explore (expand 2-hop + select + fit)	v3	300 µs	357 µs	977 µs	279 µs	1858
scn: explore (expand 2-hop + select + fit)	gpu	37.6 µs	41.0 µs	82.3 µs	33.8 µs	16441
scn: select-all + fit (listeners on)	v3	25.3 ms	26.1 ms	29.8 ms	23.5 ms	22
scn: select-all + fit (listeners on)	gpu	969 µs	1.01 ms	1.46 ms	928 µs	664
scn: drag (select band of 100 + 8-step shift, position listener)	v3	1.62 ms	1.65 ms	2.12 ms	1.58 ms	397
scn: drag (select band of 100 + 8-step shift, position listener)	gpu	99.9 µs	108 µs	362 µs	97.3 µs	6455
scn: edit (remove + re-add 256 nodes + cascade, add/remove listeners)	v3	44.5 ms	44.7 ms	45.5 ms	43.8 ms	12
scn: edit (remove + re-add 256 nodes + cascade, add/remove listeners)	gpu	6.57 ms	6.66 ms	8.37 ms	6.01 ms	97
scn: refresh (bulk data write + mapped labels + filter + fit, data listener)	v3	7.57 ms	7.63 ms	9.11 ms	6.84 ms	83
scn: refresh (bulk data write + mapped labels + filter + fit, data listener)	gpu	4.23 ms	4.38 ms	6.95 ms	3.87 ms	147

traversal N=2,000 · 2.0 minv4 (gpu) v3

trav: neighborhood()

3.9×

trav: closedNeighborhood()

2.6×

trav: connectedEdges()

2.5×

trav: outgoers()

4.9×

trav: incomers()

4.9×

trav: successors()

719×

trav: predecessors()

723×

trav: band neighborhood()

5.0×

trav: band connectedEdges()

1.46×

trav: band roots()

24×

trav: band sources()

1.48×

trav: band connectedNodes()

2.8×

100 ns1 µs10 µs100 µs1 ms10 ms100 ms1 s

All stats (table view)

group	bench	p50	avg	p99	min	samples
trav: neighborhood()	v3	3.97 µs	4.22 µs	5.28 µs	3.48 µs	36
trav: neighborhood()	gpu	1.03 µs	1.07 µs	2.05 µs	863 ns	154
trav: closedNeighborhood()	v3	5.81 µs	5.68 µs	6.23 µs	5.15 µs	26
trav: closedNeighborhood()	gpu	2.20 µs	2.31 µs	2.78 µs	2.12 µs	69
trav: connectedEdges()	v3	1.01 µs	1.09 µs	1.81 µs	977 ns	153
trav: connectedEdges()	gpu	401 ns	414 ns	700 ns	392 ns	407
trav: outgoers()	v3	2.29 µs	2.44 µs	2.93 µs	2.22 µs	65
trav: outgoers()	gpu	465 ns	545 ns	2.03 µs	440 ns	305
trav: incomers()	v3	2.31 µs	2.44 µs	3.03 µs	2.22 µs	66
trav: incomers()	gpu	472 ns	549 ns	2.02 µs	443 ns	303
trav: successors()	v3	431 ms	433 ms	443 ms	426 ms	12
trav: successors()	gpu	599 µs	636 µs	1.89 ms	576 µs	1065
trav: predecessors()	v3	431 ms	434 ms	443 ms	427 ms	12
trav: predecessors()	gpu	595 µs	629 µs	1.84 ms	578 µs	1079
trav: band neighborhood()	v3	184 µs	201 µs	393 µs	179 µs	3464
trav: band neighborhood()	gpu	36.9 µs	37.7 µs	40.9 µs	36.4 µs	12
trav: band connectedEdges()	v3	24.0 µs	24.6 µs	27.1 µs	23.8 µs	12
trav: band connectedEdges()	gpu	16.5 µs	17.3 µs	19.7 µs	16.3 µs	12
trav: band roots()	v3	109 µs	118 µs	186 µs	103 µs	5910
trav: band roots()	gpu	4.54 µs	4.56 µs	4.66 µs	4.49 µs	34
trav: band sources()	v3	6.48 µs	6.45 µs	6.68 µs	6.20 µs	23
trav: band sources()	gpu	4.37 µs	4.40 µs	4.56 µs	4.25 µs	34
trav: band connectedNodes()	v3	14.0 µs	14.0 µs	14.1 µs	13.9 µs	9
trav: band connectedNodes()	gpu	4.96 µs	4.99 µs	5.12 µs	4.91 µs	30

algorithms N=2,000 · 4.6 minv4 (gpu) v3

algo: bfs (whole graph)

58×

algo: dfs (whole graph)

59×

algo: dijkstra + pathTo

56×

algo: aStar

3.4×

algo: bellmanFord

20×

algo: kruskal

28×

algo: tarjan SCC

33×

algo: hopcroft-tarjan biconnected

24×

algo: hierholzer (directed)

2.9×

algo: betweennessCentrality (unweighted)

11×

algo: degreeCentralityNormalized

37×

algo: closenessCentrality (one root)

70×

algo: degreeCentrality (one root)

34×

100 ns1 µs10 µs100 µs1 ms10 ms100 ms1 s10 s

All stats (table view)

group	bench	p50	avg	p99	min	samples
algo: bfs (whole graph)	v3	50.2 ms	53.2 ms	57.6 ms	49.4 ms	9
algo: bfs (whole graph)	gpu	866 µs	891 µs	1.38 ms	844 µs	781
algo: dfs (whole graph)	v3	49.0 ms	49.5 ms	51.7 ms	48.1 ms	10
algo: dfs (whole graph)	gpu	828 µs	852 µs	1.38 ms	792 µs	818
algo: dijkstra + pathTo	v3	52.6 ms	53.6 ms	55.8 ms	51.7 ms	10
algo: dijkstra + pathTo	gpu	938 µs	955 µs	1.23 ms	893 µs	729
algo: aStar	v3	1.65 ms	1.70 ms	2.38 ms	1.58 ms	405
algo: aStar	gpu	489 µs	505 µs	796 µs	475 µs	1381
algo: bellmanFord	v3	231 ms	226 ms	244 ms	204 ms	12
algo: bellmanFord	gpu	11.7 ms	11.8 ms	12.3 ms	11.6 ms	56
algo: kruskal	v3	13.0 ms	14.0 ms	16.9 ms	12.5 ms	45
algo: kruskal	gpu	462 µs	483 µs	950 µs	449 µs	1443
algo: tarjan SCC	v3	22.4 ms	23.4 ms	27.2 ms	21.2 ms	26
algo: tarjan SCC	gpu	673 µs	816 µs	4.90 ms	627 µs	846
algo: hopcroft-tarjan biconnected	v3	84.9 ms	85.8 ms	89.7 ms	80.9 ms	12
algo: hopcroft-tarjan biconnected	gpu	3.53 ms	3.65 ms	4.62 ms	3.44 ms	185
algo: hierholzer (directed)	v3	22.4 ms	23.3 ms	28.6 ms	21.3 ms	26
algo: hierholzer (directed)	gpu	7.82 ms	8.09 ms	12.2 ms	7.58 ms	77
algo: betweennessCentrality (unweighted)	v3	4.04 s	4.06 s	4.13 s	4.00 s	12
algo: betweennessCentrality (unweighted)	gpu	360 ms	376 ms	442 ms	346 ms	12
algo: degreeCentralityNormalized	v3	14.4 ms	15.4 ms	20.0 ms	13.6 ms	41
algo: degreeCentralityNormalized	gpu	389 µs	405 µs	807 µs	376 µs	1718
algo: closenessCentrality (one root)	v3	59.3 ms	61.4 ms	67.9 ms	58.1 ms	12
algo: closenessCentrality (one root)	gpu	853 µs	884 µs	1.40 ms	828 µs	783
algo: degreeCentrality (one root)	v3	7.40 µs	7.26 µs	7.74 µs	6.53 µs	20
algo: degreeCentrality (one root)	gpu	218 ns	225 ns	290 ns	213 ns	751

algorithms N=500 · 10.4 minv4 (gpu) v3

algo: bfs (whole graph)

57×

algo: dfs (whole graph)

56×

algo: dijkstra + pathTo

49×

algo: aStar

3.5×

algo: bellmanFord

20×

algo: kruskal

25×

algo: tarjan SCC

34×

algo: hopcroft-tarjan biconnected

26×

algo: hierholzer (directed)

5.2×

algo: betweennessCentrality (unweighted)

12×

algo: degreeCentralityNormalized

37×

algo: closenessCentrality (one root)

69×

algo: degreeCentrality (one root)

35×

algo: pageRank (20 iters)

70×

algo: closenessCentrality (weighted, one root)

53×

algo: floydWarshall

1.93×

algo: closenessCentralityNormalized

1.78×

algo: markovClustering

1.11×

algo: hierarchicalClustering (threshold)

8.3×

algo: kMeans (fixed centroids)

4.6×

algo: kMedoids

8.7×

algo: fuzzyCMeans

11×

algo: affinityPropagation

5.1×

algo: kargerStein

1.04×

algo: betweennessCentrality (weighted)

8.8×

100 ns1 µs10 µs100 µs1 ms10 ms100 ms1 s10 s

All stats (table view)

group	bench	p50	avg	p99	min	samples
algo: bfs (whole graph)	v3	12.7 ms	13.1 ms	17.5 ms	10.7 ms	49
algo: bfs (whole graph)	gpu	224 µs	233 µs	387 µs	206 µs	2943
algo: dfs (whole graph)	v3	11.4 ms	12.3 ms	15.1 ms	11.1 ms	53
algo: dfs (whole graph)	gpu	204 µs	214 µs	348 µs	200 µs	3261
algo: dijkstra + pathTo	v3	11.8 ms	12.5 ms	15.5 ms	11.3 ms	52
algo: dijkstra + pathTo	gpu	240 µs	254 µs	414 µs	230 µs	2744
algo: aStar	v3	443 µs	453 µs	648 µs	431 µs	1545
algo: aStar	gpu	127 µs	132 µs	184 µs	123 µs	5311
algo: bellmanFord	v3	16.8 ms	17.0 ms	17.9 ms	16.5 ms	37
algo: bellmanFord	gpu	824 µs	888 µs	1.99 ms	805 µs	782
algo: kruskal	v3	3.25 ms	3.58 ms	7.26 ms	3.12 ms	188
algo: kruskal	gpu	132 µs	143 µs	268 µs	125 µs	4805
algo: tarjan SCC	v3	5.34 ms	5.75 ms	9.38 ms	5.21 ms	118
algo: tarjan SCC	gpu	159 µs	175 µs	316 µs	151 µs	3948
algo: hopcroft-tarjan biconnected	v3	20.9 ms	21.6 ms	24.0 ms	19.4 ms	29
algo: hopcroft-tarjan biconnected	gpu	799 µs	818 µs	1.14 ms	776 µs	851
algo: hierholzer (directed)	v3	4.28 ms	4.60 ms	7.74 ms	4.13 ms	142
algo: hierholzer (directed)	gpu	821 µs	852 µs	1.22 ms	794 µs	798
algo: betweennessCentrality (unweighted)	v3	253 ms	255 ms	258 ms	250 ms	12
algo: betweennessCentrality (unweighted)	gpu	20.5 ms	20.6 ms	21.0 ms	20.2 ms	29
algo: degreeCentralityNormalized	v3	3.52 ms	3.79 ms	7.18 ms	3.40 ms	178
algo: degreeCentralityNormalized	gpu	94.8 µs	99.2 µs	170 µs	91.0 µs	7036
algo: closenessCentrality (one root)	v3	14.3 ms	15.1 ms	18.4 ms	13.1 ms	42
algo: closenessCentrality (one root)	gpu	206 µs	216 µs	348 µs	202 µs	3233
algo: degreeCentrality (one root)	v3	7.60 µs	7.58 µs	8.36 µs	6.88 µs	19
algo: degreeCentrality (one root)	gpu	216 ns	224 ns	326 ns	210 ns	757
algo: pageRank (20 iters)	v3	2.81 ms	3.01 ms	6.29 ms	2.72 ms	225
algo: pageRank (20 iters)	gpu	39.9 µs	44.7 µs	162 µs	38.7 µs	15517
algo: closenessCentrality (weighted, one root)	v3	13.8 ms	14.8 ms	18.0 ms	13.2 ms	43
algo: closenessCentrality (weighted, one root)	gpu	259 µs	279 µs	496 µs	249 µs	2508
algo: floydWarshall	v3	249 ms	273 ms	386 ms	244 ms	12
algo: floydWarshall	gpu	129 ms	133 ms	138 ms	127 ms	12
algo: closenessCentralityNormalized	v3	277 ms	279 ms	286 ms	272 ms	12
algo: closenessCentralityNormalized	gpu	155 ms	174 ms	238 ms	153 ms	12
algo: markovClustering	v3	4.17 s	4.39 s	4.68 s	4.08 s	12
algo: markovClustering	gpu	3.76 s	3.76 s	3.78 s	3.75 s	12
algo: hierarchicalClustering (threshold)	v3	34.2 ms	34.2 ms	34.9 ms	33.5 ms	17
algo: hierarchicalClustering (threshold)	gpu	4.12 ms	4.28 ms	5.65 ms	4.03 ms	158
algo: kMeans (fixed centroids)	v3	8.94 ms	9.02 ms	9.75 ms	8.73 ms	74
algo: kMeans (fixed centroids)	gpu	1.93 ms	1.98 ms	3.10 ms	1.87 ms	350
algo: kMedoids	v3	358 ms	310 ms	391 ms	142 ms	12
algo: kMedoids	gpu	41.2 ms	39.9 ms	47.4 ms	26.5 ms	15
algo: fuzzyCMeans	v3	79.3 ms	80.5 ms	84.0 ms	77.2 ms	12
algo: fuzzyCMeans	gpu	6.98 ms	7.03 ms	7.67 ms	6.68 ms	95
algo: affinityPropagation	v3	209 ms	211 ms	218 ms	200 ms	12
algo: affinityPropagation	gpu	40.8 ms	59.3 ms	115 ms	40.3 ms	12
algo: kargerStein	v3	318 ms	323 ms	332 ms	312 ms	12
algo: kargerStein	gpu	307 ms	309 ms	317 ms	298 ms	12
algo: betweennessCentrality (weighted)	v3	718 ms	720 ms	729 ms	711 ms	12
algo: betweennessCentrality (weighted)	gpu	81.3 ms	82.7 ms	82.4 ms	79.9 ms	12

mappers N=2,000 · 23.9 s

mapper: bulk write, color — cpu eval

1.31 ms

mapper: bulk write, color — gpu path

175 µs

mapper: bulk write, color + size — cpu eval

1.60 ms

mapper: bulk write, color + size — gpu color / cpu size

1.50 ms

mapper: bulk write, mapped labels (round-5 path) — label refresh

4.06 ms

100 µs1 ms10 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
mapper: bulk write, color	cpu eval	1.31 ms	1.36 ms	1.99 ms	1.27 ms	513
mapper: bulk write, color	gpu path	175 µs	185 µs	320 µs	170 µs	3793
mapper: bulk write, color + size	cpu eval	1.60 ms	1.63 ms	2.00 ms	1.54 ms	425
mapper: bulk write, color + size	gpu color / cpu size	1.50 ms	1.50 ms	1.70 ms	1.43 ms	465
mapper: bulk write, mapped labels (round-5 path)	label refresh	4.06 ms	4.30 ms	6.93 ms	3.48 ms	158

layouts N=2,000 · 1.5 minv4 (gpu) v3

94×

19×

60×

12×

28×

36×

11×

114 ms

layout: force seed — spectral vs scatter (20 iterations) — spectral (landmark MDS, the default)

112 ms

56.5 ms

356 µs

31.5 µs

521 ns

layout: contract fixed cost (empty impl vs bulk placement) — wrapper + bulk write

39.4 µs

10×

100 ns1 µs10 µs100 µs1 ms10 ms100 ms1 s

All stats (table view)

group	bench	p50	avg	p99	min	samples
layout: grid	v3	12.1 ms	12.5 ms	14.9 ms	10.6 ms	51
layout: grid	gpu	129 µs	143 µs	615 µs	80.4 µs	4649
layout: preset	v3	5.75 ms	6.03 ms	7.89 ms	5.35 ms	104
layout: preset	gpu	295 µs	325 µs	718 µs	287 µs	2130
layout: preset (fn form)	v3	5.78 ms	6.05 ms	8.82 ms	5.24 ms	111
layout: preset (fn form)	gpu	96.3 µs	103 µs	352 µs	93.5 µs	6508
layout: circle	v3	9.65 ms	10.2 ms	15.0 ms	9.00 ms	64
layout: circle	gpu	816 µs	944 µs	2.89 ms	803 µs	714
layout: concentric	v3	31.6 ms	31.3 ms	36.2 ms	27.2 ms	18
layout: concentric	gpu	1.15 ms	1.32 ms	4.83 ms	1.11 ms	512
layout: breadthfirst	v3	388 ms	399 ms	435 ms	382 ms	12
layout: breadthfirst	gpu	10.8 ms	11.2 ms	15.9 ms	10.2 ms	55
layout: random	v3	5.76 ms	6.09 ms	9.39 ms	5.17 ms	109
layout: random	gpu	537 µs	603 µs	1.55 ms	525 µs	1106
layout: force (CPU executor, 20 iterations)	gpu	114 ms	129 ms	143 ms	109 ms	12
layout: force seed — spectral vs scatter (20 iterations)	spectral (landmark MDS, the default)	112 ms	113 ms	121 ms	103 ms	12
layout: force seed — spectral vs scatter (20 iterations)	scatter	56.5 ms	60.7 ms	75.2 ms	55.3 ms	12
layout: plumbing (layoutPositions vs setPositions)	finisher	356 µs	389 µs	912 µs	348 µs	1779
layout: plumbing (layoutPositions vs setPositions)	bulk	31.5 µs	45.1 µs	259 µs	29.3 µs	14800
layout: contract fixed cost (empty impl vs bulk placement)	wrapper only	521 ns	980 ns	3.86 µs	510 ns	161
layout: contract fixed cost (empty impl vs bulk placement)	wrapper + bulk write	39.4 µs	41.8 µs	50.1 µs	37.8 µs	12
layout: scope (eles.layout, 10% of nodes)	v3	947 µs	1.01 ms	2.33 ms	905 µs	668
layout: scope (eles.layout, 10% of nodes)	gpu	94.1 µs	106 µs	262 µs	92.2 µs	6522

style N=2,000 · 1.7 minv4 (gpu) v3

style: sheet swap (compile + applyAll)

5.8×

style: compile vs compile + apply — compile only (in batch)

24.6 µs

style: compile vs compile + apply — compile + apply

2.24 ms

style: first apply on add (256-node band)

1.94×

style: applyAll, flat vs compound (100 parents — the 14.6 partition) — flat

2.26 ms

style: applyAll, flat vs compound (100 parents — the 14.6 partition) — compound

2.72 ms

style: applyAll — constant vs state-case (the 57.1d partition) — constant sheet

2.25 ms

style: applyAll — constant vs state-case (the 57.1d partition) — state-case sheet

2.26 ms

style: applyAll — state-case vs data-case (what the partition avoids) — state-case sheet

2.24 ms

style: applyAll — state-case vs data-case (what the partition avoids) — data-case sheet

10.0 ms

style: select+unselect 256-band — constant vs state-case sheet — constant sheet (skips restyle)

6.77 µs

style: select+unselect 256-band — constant vs state-case sheet — paint-only state sheet (round-61 diff path)

32.9 µs

style: select+unselect 256-band — constant vs state-case sheet — state-case sheet (all channels; full write)

372 µs

style: read one prop (color)

1.73×

style: read one prop (width)

2.2×

style: numericStyle

2.9×

style: renderedStyle (one prop)

13×

style: whole-object style()

6.2×

style: bypass set (ele.style name,value)

4.1×

style: bypass set + removeStyle round-trip

5.8×

style: select+unselect 256-band under the default sheet — bypass punch-out — no bypasses (diff path; the O(0) gate row)

33.7 µs

style: select+unselect 256-band under the default sheet — bypass punch-out — 128 of 256 bypassed (punched out to merged writes)

226 µs

style: sheet swap (applyAll) — 0 vs 128 bypasses — no bypasses

2.43 ms

style: sheet swap (applyAll) — 0 vs 128 bypasses — 128 bypassed

8.90 ms

10 ns100 ns1 µs10 µs100 µs1 ms10 ms100 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
style: sheet swap (compile + applyAll)	v3	13.0 ms	13.5 ms	16.5 ms	12.5 ms	46
style: sheet swap (compile + applyAll)	gpu	2.25 ms	2.29 ms	2.67 ms	2.19 ms	289
style: compile vs compile + apply	compile only (in batch)	24.6 µs	27.2 µs	36.2 µs	23.8 µs	12
style: compile vs compile + apply	compile + apply	2.24 ms	2.28 ms	2.71 ms	2.15 ms	305
style: first apply on add (256-node band)	v3	1.41 ms	1.68 ms	7.11 ms	1.16 ms	403
style: first apply on add (256-node band)	gpu	726 µs	865 µs	2.62 ms	698 µs	798
style: applyAll, flat vs compound (100 parents — the 14.6 partition)	flat	2.26 ms	2.35 ms	3.28 ms	2.18 ms	295
style: applyAll, flat vs compound (100 parents — the 14.6 partition)	compound	2.72 ms	2.81 ms	4.70 ms	2.62 ms	244
style: applyAll — constant vs state-case (the 57.1d partition)	constant sheet	2.25 ms	2.32 ms	3.19 ms	2.18 ms	299
style: applyAll — constant vs state-case (the 57.1d partition)	state-case sheet	2.26 ms	2.30 ms	2.91 ms	2.19 ms	301
style: applyAll — state-case vs data-case (what the partition avoids)	state-case sheet	2.24 ms	2.29 ms	2.82 ms	2.20 ms	301
style: applyAll — state-case vs data-case (what the partition avoids)	data-case sheet	10.0 ms	10.1 ms	11.1 ms	9.67 ms	62
style: select+unselect 256-band — constant vs state-case sheet	constant sheet (skips restyle)	6.77 µs	6.80 µs	6.97 µs	6.68 µs	20
style: select+unselect 256-band — constant vs state-case sheet	paint-only state sheet (round-61 diff path)	32.9 µs	34.8 µs	75.1 µs	31.8 µs	19130
style: select+unselect 256-band — constant vs state-case sheet	state-case sheet (all channels; full write)	372 µs	389 µs	660 µs	368 µs	1798
style: read one prop (color)	v3	95.0 ns	106 ns	624 ns	87.0 ns	2006561
style: read one prop (color)	gpu	54.8 ns	59.4 ns	104 ns	53.2 ns	2868
style: read one prop (width)	v3	115 ns	126 ns	435 ns	100 ns	1868713
style: read one prop (width)	gpu	51.9 ns	56.1 ns	302 ns	50.7 ns	3032
style: numericStyle	v3	141 ns	133 ns	382 ns	87.0 ns	1784946
style: numericStyle	gpu	48.9 ns	52.0 ns	117 ns	47.6 ns	3282
style: renderedStyle (one prop)	v3	883 ns	949 ns	1.98 µs	844 ns	556273
style: renderedStyle (one prop)	gpu	65.6 ns	70.0 ns	225 ns	62.7 ns	2441
style: whole-object style()	v3	85.5 µs	91.3 µs	187 µs	83.8 µs	7610
style: whole-object style()	gpu	13.8 µs	15.2 µs	49.6 µs	13.0 µs	44410
style: bypass set (ele.style name,value)	v3	10.0 µs	10.1 µs	10.7 µs	9.69 µs	11
style: bypass set (ele.style name,value)	gpu	2.46 µs	2.76 µs	4.13 µs	2.36 µs	55
style: bypass set + removeStyle round-trip	v3	21.0 µs	23.3 µs	50.0 µs	20.1 µs	29070
style: bypass set + removeStyle round-trip	gpu	3.60 µs	3.62 µs	4.43 µs	3.18 µs	42
style: select+unselect 256-band under the default sheet — bypass punch-out	no bypasses (diff path; the O(0) gate row)	33.7 µs	34.9 µs	65.9 µs	32.7 µs	18977
style: select+unselect 256-band under the default sheet — bypass punch-out	128 of 256 bypassed (punched out to merged writes)	226 µs	237 µs	416 µs	219 µs	2942
style: sheet swap (applyAll) — 0 vs 128 bypasses	no bypasses	2.43 ms	2.47 ms	3.06 ms	2.33 ms	265
style: sheet swap (applyAll) — 0 vs 128 bypasses	128 bypassed	8.90 ms	9.06 ms	10.4 ms	8.62 ms	73

style-bundle N=2,000 · 58.6 sv4 (gpu) v3

style( 'background-color' ) (was #4 in the switch)

1.12×

style( 'border-width' ) (was #6 in the switch)

1.15×

style( 'text-wrap' ) (was #73 in the switch)

1.70×

style( 'text-max-width' ) (was #74 in the switch)

1.73×

style( 'taxi-radius' ) (was #142 in the switch)

1.29×

style( 'target-distance-from-node' ) (was #150 in the switch)

1.27×

whole-object style() (gpu-only: v3 returns a different shape) — gpu node

13.4 µs

whole-object style() (gpu-only: v3 returns a different shape) — gpu edge

14.4 µs

colour-valued vs numeric reads (the string build) — gpu background-color (colour)

73.8 ns

colour-valued vs numeric reads (the string build) — gpu border-color (colour)

65.7 ns

colour-valued vs numeric reads (the string build) — gpu border-width (number)

67.1 ns

colour-valued vs numeric reads (the string build) — gpu width (number)

60.9 ns

the getters that ride readProp — gpu numericStyle( width )

62.9 ns

the getters that ride readProp — gpu effectiveOpacity()

20.6 ns

the getters that ride readProp — gpu renderedStyle( width )

77.2 ns

10 ns100 ns1 µs10 µs100 µs

All stats (table view)

group	bench	p50	avg	p99	min	samples
style( 'background-color' ) (was #4 in the switch)	v3	59.7 ns	61.8 ns	99.9 ns	56.8 ns	2775
style( 'background-color' ) (was #4 in the switch)	gpu	53.4 ns	56.2 ns	96.2 ns	52.2 ns	3043
style( 'border-width' ) (was #6 in the switch)	v3	61.5 ns	62.9 ns	92.0 ns	58.7 ns	2728
style( 'border-width' ) (was #6 in the switch)	gpu	53.7 ns	57.0 ns	116 ns	52.0 ns	2997
style( 'text-wrap' ) (was #73 in the switch)	v3	60.1 ns	62.2 ns	97.1 ns	58.8 ns	2751
style( 'text-wrap' ) (was #73 in the switch)	gpu	35.3 ns	36.1 ns	47.9 ns	34.1 ns	4739
style( 'text-max-width' ) (was #74 in the switch)	v3	59.5 ns	63.1 ns	128 ns	57.7 ns	2723
style( 'text-max-width' ) (was #74 in the switch)	gpu	34.5 ns	35.0 ns	46.1 ns	33.1 ns	4898
style( 'taxi-radius' ) (was #142 in the switch)	v3	63.4 ns	65.4 ns	89.5 ns	59.7 ns	2627
style( 'taxi-radius' ) (was #142 in the switch)	gpu	49.3 ns	55.4 ns	99.2 ns	47.9 ns	3071
style( 'target-distance-from-node' ) (was #150 in the switch)	v3	61.4 ns	63.7 ns	92.2 ns	59.2 ns	2695
style( 'target-distance-from-node' ) (was #150 in the switch)	gpu	48.4 ns	50.0 ns	96.5 ns	47.1 ns	3423
whole-object style() (gpu-only: v3 returns a different shape)	gpu node	13.4 µs	14.7 µs	27.4 µs	12.7 µs	45818
whole-object style() (gpu-only: v3 returns a different shape)	gpu edge	14.4 µs	15.6 µs	36.5 µs	13.8 µs	43075
colour-valued vs numeric reads (the string build)	gpu background-color (colour)	73.8 ns	76.6 ns	119 ns	71.5 ns	2232
colour-valued vs numeric reads (the string build)	gpu border-color (colour)	65.7 ns	68.3 ns	104 ns	65.1 ns	2505
colour-valued vs numeric reads (the string build)	gpu border-width (number)	67.1 ns	68.9 ns	93.0 ns	65.8 ns	2483
colour-valued vs numeric reads (the string build)	gpu width (number)	60.9 ns	63.7 ns	107 ns	58.7 ns	2680
the getters that ride readProp	gpu numericStyle( width )	62.9 ns	65.6 ns	103 ns	62.0 ns	2610
the getters that ride readProp	gpu effectiveOpacity()	20.6 ns	21.3 ns	34.2 ns	19.4 ns	8046
the getters that ride readProp	gpu renderedStyle( width )	77.2 ns	80.3 ns	126 ns	73.9 ns	2128

load N=2,000 · 58.6 sv4 (gpu) v3

load: init from the definition form (construct + dispose)

26×

load: v4 ingest forms (construct + dispose) — definition form

4.58 ms

load: v4 ingest forms (construct + dispose) — columnar form

3.26 ms

load: v4 ingest forms (construct + dispose) — wire buffer

2.86 ms

load: v4 ingest forms (construct + dispose) — def clone alone (control)

301 µs

load: conversion — toColumnarElements

693 µs

load: conversion — serializeElements (from columnar)

179 µs

load: conversion — serializeElements (from defs)

880 µs

load: conversion — deserializeElements

4.54 µs

load: export from a live graph — cy.serialize() (wire)

501 µs

load: export from a live graph — cy.json() (v4)

727 µs

load: cy.json()

2.8×

load: add a 256-node band (add + remove)

1.66×

load: add a 256-node band, v4 forms — definition form

2.26 ms

load: add a 256-node band, v4 forms — columnar form

1.71 ms

load: add a 256-node band, v4 forms — wire buffer

1.95 ms

1 µs10 µs100 µs1 ms10 ms100 ms1 s

All stats (table view)

group	bench	p50	avg	p99	min	samples
load: init from the definition form (construct + dispose)	v3	135 ms	147 ms	175 ms	126 ms	12
load: init from the definition form (construct + dispose)	gpu	5.19 ms	7.00 ms	20.8 ms	4.28 ms	88
load: v4 ingest forms (construct + dispose)	definition form	4.58 ms	6.56 ms	16.8 ms	4.24 ms	102
load: v4 ingest forms (construct + dispose)	columnar form	3.26 ms	4.49 ms	13.3 ms	3.20 ms	152
load: v4 ingest forms (construct + dispose)	wire buffer	2.86 ms	4.18 ms	12.5 ms	2.73 ms	162
load: v4 ingest forms (construct + dispose)	def clone alone (control)	301 µs	592 µs	8.37 ms	208 µs	1113
load: conversion	toColumnarElements	693 µs	745 µs	1.29 ms	673 µs	928
load: conversion	serializeElements (from columnar)	179 µs	360 µs	956 µs	172 µs	1934
load: conversion	serializeElements (from defs)	880 µs	955 µs	1.83 ms	856 µs	728
load: conversion	deserializeElements	4.54 µs	4.44 µs	5.95 µs	3.41 µs	29
load: export from a live graph	cy.serialize() (wire)	501 µs	571 µs	1.96 ms	486 µs	1204
load: export from a live graph	cy.json() (v4)	727 µs	1.04 ms	6.38 ms	707 µs	642
load: cy.json()	v3	2.05 ms	2.38 ms	5.30 ms	1.98 ms	285
load: cy.json()	gpu	738 µs	813 µs	2.83 ms	706 µs	856
load: add a 256-node band (add + remove)	v3	3.67 ms	4.40 ms	10.8 ms	3.28 ms	149
load: add a 256-node band (add + remove)	gpu	2.21 ms	2.45 ms	7.77 ms	1.49 ms	275
load: add a 256-node band, v4 forms	definition form	2.26 ms	2.35 ms	5.14 ms	1.50 ms	291
load: add a 256-node band, v4 forms	columnar form	1.71 ms	1.83 ms	3.50 ms	1.26 ms	339
load: add a 256-node band, v4 forms	wire buffer	1.95 ms	2.04 ms	3.82 ms	1.39 ms	336

spatial N=2,000 · 1.9 minv4 (gpu) v3

pick: hit (the walk; one inside-test) — zoom 1

20.6 µs

pick: hit (the walk; one inside-test) — zoom 2

21.2 µs

pick: miss walk — exact vs 24 px touch halo (57.9) — exact

45.8 µs

pick: miss walk — exact vs 24 px touch halo (57.9) — padPx 24

44.5 µs

pick: shape inside-test, 2000 coincident misses — ellipse

89.0 µs

pick: shape inside-test, 2000 coincident misses — star

235 µs

pick: shape inside-test, 2000 coincident misses — round-hexagon

797 µs

pick: shape inside-test, 2000 coincident misses — barrel

681 µs

pick: shape inside-test, 2000 coincident misses — cut-rectangle

100 µs

pick: shape inside-test, 2000 coincident misses — polygon

212 µs

pick: round-hexagon inside-test by zoom (not affine-invariant) — zoom 1

800 µs

pick: round-hexagon inside-test by zoom (not affine-invariant) — zoom 2

790 µs

pick: background miss (full highWater walk) — gpu

45.2 µs

pick: label box in the scan (text-events) — text-events: no

45.1 µs

pick: label box in the scan (text-events) — text-events: yes

126 µs

box: elementsInBox by covered fraction — 10%

149 µs

box: elementsInBox by covered fraction — 50%

196 µs

box: elementsInBox by covered fraction — 100%

321 µs

box: label containment (boxSelectionIncludesLabels) — off

230 µs

box: label containment (boxSelectionIncludesLabels) — on

262 µs

box: overlap vs contain (straight edges) — contain

234 µs

box: overlap vs contain (straight edges) — overlap

485 µs

box: overlap vs contain (curved edges — the segment walk) — contain

1.28 ms

box: overlap vs contain (curved edges — the segment walk) — overlap

2.18 ms

bounds: whole-graph boundingBox()

5.9×

bounds: boundingBox, labels in vs out (v4) — includeLabels: true (default)

1.34 ms

bounds: boundingBox, labels in vs out (v4) — includeLabels: false

816 µs

bounds: cy.fit() (whole-graph scan)

33×

bounds: getFitViewport() (compute without committing)

33×

bounds: boundingBoxAt (v4 only) — gpu

2.63 ms

bounds: one node boundingBox()

1.61×

bounds: labelBoundingBox (v4 only) — gpu

429 ns

bounds: compound + taxi conservative scan (round 54, v4 only) — warm (curve bbs memoized)

76.7 µs

bounds: compound + taxi conservative scan (round 54, v4 only) — cold (geometry write per call)

151 µs

100 ns1 µs10 µs100 µs1 ms10 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
pick: hit (the walk; one inside-test)	zoom 1	20.6 µs	21.4 µs	38.9 µs	20.4 µs	32176
pick: hit (the walk; one inside-test)	zoom 2	21.2 µs	21.5 µs	21.9 µs	20.7 µs	12
pick: miss walk — exact vs 24 px touch halo (57.9)	exact	45.8 µs	46.3 µs	46.9 µs	45.5 µs	12
pick: miss walk — exact vs 24 px touch halo (57.9)	padPx 24	44.5 µs	46.2 µs	88.1 µs	43.2 µs	15019
pick: shape inside-test, 2000 coincident misses	ellipse	89.0 µs	92.7 µs	169 µs	84.1 µs	7531
pick: shape inside-test, 2000 coincident misses	star	235 µs	245 µs	317 µs	226 µs	2855
pick: shape inside-test, 2000 coincident misses	round-hexagon	797 µs	809 µs	975 µs	779 µs	865
pick: shape inside-test, 2000 coincident misses	barrel	681 µs	697 µs	848 µs	660 µs	1002
pick: shape inside-test, 2000 coincident misses	cut-rectangle	100 µs	102 µs	142 µs	97.5 µs	6872
pick: shape inside-test, 2000 coincident misses	polygon	212 µs	216 µs	264 µs	208 µs	3243
pick: round-hexagon inside-test by zoom (not affine-invariant)	zoom 1	800 µs	812 µs	1.13 ms	769 µs	862
pick: round-hexagon inside-test by zoom (not affine-invariant)	zoom 2	790 µs	809 µs	1.01 ms	764 µs	865
pick: background miss (full highWater walk)	gpu	45.2 µs	45.7 µs	46.5 µs	44.6 µs	12
pick: label box in the scan (text-events)	text-events: no	45.1 µs	45.5 µs	46.1 µs	44.7 µs	12
pick: label box in the scan (text-events)	text-events: yes	126 µs	130 µs	216 µs	121 µs	5343
box: elementsInBox by covered fraction	10%	149 µs	154 µs	256 µs	148 µs	4530
box: elementsInBox by covered fraction	50%	196 µs	205 µs	356 µs	191 µs	3371
box: elementsInBox by covered fraction	100%	321 µs	336 µs	790 µs	312 µs	2065
box: label containment (boxSelectionIncludesLabels)	off	230 µs	240 µs	478 µs	225 µs	2911
box: label containment (boxSelectionIncludesLabels)	on	262 µs	277 µs	631 µs	254 µs	2520
box: overlap vs contain (straight edges)	contain	234 µs	244 µs	494 µs	231 µs	2863
box: overlap vs contain (straight edges)	overlap	485 µs	504 µs	999 µs	432 µs	1375
box: overlap vs contain (curved edges — the segment walk)	contain	1.28 ms	1.31 ms	1.85 ms	1.25 ms	532
box: overlap vs contain (curved edges — the segment walk)	overlap	2.18 ms	2.23 ms	2.80 ms	2.14 ms	311
bounds: whole-graph boundingBox()	v3	7.76 ms	7.94 ms	9.41 ms	7.45 ms	84
bounds: whole-graph boundingBox()	gpu	1.32 ms	1.37 ms	1.86 ms	1.29 ms	506
bounds: boundingBox, labels in vs out (v4)	includeLabels: true (default)	1.34 ms	1.38 ms	2.04 ms	1.31 ms	503
bounds: boundingBox, labels in vs out (v4)	includeLabels: false	816 µs	842 µs	1.21 ms	778 µs	829
bounds: cy.fit() (whole-graph scan)	v3	7.73 ms	8.03 ms	10.4 ms	7.51 ms	83
bounds: cy.fit() (whole-graph scan)	gpu	237 µs	247 µs	342 µs	233 µs	2824
bounds: getFitViewport() (compute without committing)	v3	7.70 ms	7.92 ms	9.50 ms	7.33 ms	83
bounds: getFitViewport() (compute without committing)	gpu	236 µs	247 µs	430 µs	229 µs	2832
bounds: boundingBoxAt (v4 only)	gpu	2.63 ms	2.71 ms	3.50 ms	2.62 ms	247
bounds: one node boundingBox()	v3	1.68 µs	1.70 µs	2.17 µs	1.63 µs	94
bounds: one node boundingBox()	gpu	1.04 µs	1.08 µs	1.76 µs	993 ns	152
bounds: labelBoundingBox (v4 only)	gpu	429 ns	456 ns	1.09 µs	419 ns	365
bounds: compound + taxi conservative scan (round 54, v4 only)	warm (curve bbs memoized)	76.7 µs	80.5 µs	190 µs	76.0 µs	8661
bounds: compound + taxi conservative scan (round 54, v4 only)	cold (geometry write per call)	151 µs	157 µs	294 µs	147 µs	4441

data N=2,000 · 1.4 minv4 (gpu) v3

data: bulk write, numeric column

20×

data: bulk write, dictionary string column (4 values)

16×

data: bulk write, one new dictionary entry per pass

18×

data: bulk write, object column (plain-array fallback)

21×

data: removeData (whole collection)

8.4×

data: read one numeric key

1.41×

data: read one string key (dictionary decode)

1.35×

data: read the whole object

2.2×

query: data equality ([bucket = 3] vs { data: { bucket: 3 } })

14×

query: data comparison ([weight > 3] vs { gt })

11×

query: two keys AND-ed

12×

query: membership ([bucket in ...] vs { in })

15×

query: predicate function (both sides)

6.0×

query: structural, parents (:parent vs { parent: true })

34×

query: structural, childless (:childless vs { parent: false })

13×

query: structural, orphans (:orphan vs { child: false })

32×

10 ns100 ns1 µs10 µs100 µs1 ms10 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
data: bulk write, numeric column	v3	1.67 ms	1.77 ms	3.33 ms	1.62 ms	376
data: bulk write, numeric column	gpu	84.3 µs	87.5 µs	136 µs	81.1 µs	7993
data: bulk write, dictionary string column (4 values)	v3	1.77 ms	1.81 ms	2.40 ms	1.71 ms	369
data: bulk write, dictionary string column (4 values)	gpu	109 µs	114 µs	311 µs	97.5 µs	6105
data: bulk write, one new dictionary entry per pass	v3	1.77 ms	1.91 ms	3.27 ms	1.73 ms	358
data: bulk write, one new dictionary entry per pass	gpu	99.7 µs	104 µs	164 µs	94.7 µs	6705
data: bulk write, object column (plain-array fallback)	v3	1.77 ms	1.85 ms	3.24 ms	1.70 ms	374
data: bulk write, object column (plain-array fallback)	gpu	83.3 µs	87.3 µs	152 µs	79.5 µs	8011
data: removeData (whole collection)	v3	2.76 ms	2.82 ms	3.97 ms	2.65 ms	243
data: removeData (whole collection)	gpu	328 µs	340 µs	551 µs	316 µs	2009
data: read one numeric key	v3	38.7 ns	41.3 ns	118 ns	36.7 ns	4159
data: read one numeric key	gpu	27.5 ns	28.6 ns	71.7 ns	26.7 ns	5745
data: read one string key (dictionary decode)	v3	41.9 ns	43.9 ns	97.8 ns	40.4 ns	3910
data: read one string key (dictionary decode)	gpu	31.1 ns	32.5 ns	46.8 ns	30.5 ns	5285
data: read the whole object	v3	46.2 ns	47.8 ns	76.0 ns	43.8 ns	3598
data: read the whole object	gpu	21.0 ns	22.3 ns	40.5 ns	20.5 ns	7666
query: data equality ([bucket = 3] vs { data: { bucket: 3 } })	v3	664 µs	683 µs	1.07 ms	646 µs	1022
query: data equality ([bucket = 3] vs { data: { bucket: 3 } })	gpu	46.8 µs	50.6 µs	125 µs	41.1 µs	13650
query: data comparison ([weight > 3] vs { gt })	v3	751 µs	778 µs	1.17 ms	740 µs	897
query: data comparison ([weight > 3] vs { gt })	gpu	69.9 µs	76.6 µs	221 µs	58.2 µs	9026
query: two keys AND-ed	v3	796 µs	816 µs	1.31 ms	779 µs	854
query: two keys AND-ed	gpu	65.9 µs	69.8 µs	145 µs	65.0 µs	9956
query: membership ([bucket in ...] vs { in })	v3	1.01 ms	1.07 ms	2.10 ms	980 µs	653
query: membership ([bucket in ...] vs { in })	gpu	65.6 µs	71.6 µs	140 µs	61.9 µs	9554
query: predicate function (both sides)	v3	661 µs	689 µs	1.25 ms	642 µs	1011
query: predicate function (both sides)	gpu	110 µs	114 µs	179 µs	106 µs	6136
query: structural, parents (:parent vs { parent: true })	v3	524 µs	530 µs	622 µs	512 µs	1321
query: structural, parents (:parent vs { parent: true })	gpu	15.4 µs	16.3 µs	19.3 µs	15.1 µs	12
query: structural, childless (:childless vs { parent: false })	v3	804 µs	823 µs	1.25 ms	784 µs	847
query: structural, childless (:childless vs { parent: false })	gpu	61.3 µs	79.5 µs	229 µs	56.2 µs	8643
query: structural, orphans (:orphan vs { child: false })	v3	499 µs	512 µs	787 µs	492 µs	1356
query: structural, orphans (:orphan vs { child: false })	gpu	15.4 µs	16.0 µs	15.6 µs	15.1 µs	12

events N=2,000 · 1.3 minv4 (gpu) v3

emit: position write, no listeners

28×

emit: 1 core listener

8.5×

emit: 1 element (ref-qualified) listener

7.6×

emit: delegated (v3 selector vs v4 predicate)

8.0×

emit: 10 core listeners

6.1×

emit: on() + off() round-trip

2.2×

emit: flat vs phased (compound bubbling, 14.5) — orphan (flat emit)

285 ns

emit: flat vs phased (compound bubbling, 14.5) — child of 2 ancestors (phased)

834 ns

emit: flat vs phased, no listeners — orphan (flat)

77.0 ns

emit: flat vs phased, no listeners — child (phased)

483 ns

anim: start + stop, one element

1.37×

anim: start against a running animation (round-21 eviction) — disjoint channels (composes)

1.28 µs

anim: start against a running animation (round-21 eviction) — overlapping channels (evicts)

817 ns

anim: start + stop, 512-node collection

3.0×

anim: delay() (no channels touched)

1.59×

anim: controls (v4 only) — pause + resume

458 ns

anim: controls (v4 only) — reverse

896 ns

anim: viewport animate + stop

3.0×

10 ns100 ns1 µs10 µs100 µs1 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
emit: position write, no listeners	v3	1.31 µs	1.37 µs	1.63 µs	1.26 µs	115
emit: position write, no listeners	gpu	47.3 ns	50.2 ns	108 ns	45.8 ns	3399
emit: 1 core listener	v3	1.44 µs	1.50 µs	2.20 µs	1.39 µs	102
emit: 1 core listener	gpu	169 ns	187 ns	342 ns	165 ns	860
emit: 1 element (ref-qualified) listener	v3	1.38 µs	1.46 µs	2.16 µs	1.34 µs	107
emit: 1 element (ref-qualified) listener	gpu	182 ns	201 ns	925 ns	180 ns	815
emit: delegated (v3 selector vs v4 predicate)	v3	1.49 µs	1.54 µs	2.71 µs	1.43 µs	104
emit: delegated (v3 selector vs v4 predicate)	gpu	185 ns	199 ns	353 ns	181 ns	803
emit: 10 core listeners	v3	1.93 µs	1.99 µs	2.44 µs	1.89 µs	78
emit: 10 core listeners	gpu	319 ns	327 ns	494 ns	311 ns	485
emit: on() + off() round-trip	v3	470 ns	562 ns	1.34 µs	433 ns	283
emit: on() + off() round-trip	gpu	212 ns	254 ns	959 ns	180 ns	648
emit: flat vs phased (compound bubbling, 14.5)	orphan (flat emit)	285 ns	347 ns	2.96 µs	266 ns	970078
emit: flat vs phased (compound bubbling, 14.5)	child of 2 ancestors (phased)	834 ns	885 ns	1.22 µs	796 ns	179
emit: flat vs phased, no listeners	orphan (flat)	77.0 ns	81.1 ns	156 ns	73.9 ns	1959
emit: flat vs phased, no listeners	child (phased)	483 ns	529 ns	1.88 µs	469 ns	314
anim: start + stop, one element	v3	870 ns	1.28 µs	3.42 µs	798 ns	406979
anim: start + stop, one element	gpu	633 ns	1.03 µs	4.70 µs	587 ns	474950
anim: start against a running animation (round-21 eviction)	disjoint channels (composes)	1.28 µs	1.87 µs	3.78 µs	1.07 µs	81
anim: start against a running animation (round-21 eviction)	overlapping channels (evicts)	817 ns	1.46 µs	3.83 µs	790 ns	104
anim: start + stop, 512-node collection	v3	297 µs	418 µs	1.09 ms	290 µs	1496
anim: start + stop, 512-node collection	gpu	98.5 µs	109 µs	165 µs	94.8 µs	5850
anim: delay() (no channels touched)	v3	676 ns	967 ns	3.29 µs	653 ns	165
anim: delay() (no channels touched)	gpu	426 ns	768 ns	2.85 µs	417 ns	210
anim: controls (v4 only)	pause + resume	458 ns	739 ns	2.62 µs	450 ns	207
anim: controls (v4 only)	reverse	896 ns	1.53 µs	3.47 µs	877 ns	99
anim: viewport animate + stop	v3	575 ns	820 ns	3.06 µs	559 ns	191
anim: viewport animate + stop	gpu	191 ns	214 ns	569 ns	186 ns	741

store N=2,000 · 1.0 min

id index: set x 2000 (build) — gpu

264 µs

id index: single-key operations — get (probe)

65.8 ns

id index: single-key operations — has

51.6 ns

id index: single-key operations — idAt (memoized hit)

9.02 ns

id index: single-key operations — hashAt

7.01 ns

id index: remove + re-set (churn, drives the round-11 reclaim) — gpu

232 ns

adjacency: rebuild (4000 edges, two counting passes) — gpu

39.1 µs

adjacency: reads — outDegree (O(1))

6.26 ns

adjacency: reads — outEdges

45.2 ns

adjacency: reads — connectedEdges

117 ns

adjacency: addEdge + removeEdge (the overlay path) — gpu

140 ns

blob: write x 2000 (build) — gpu

104 µs

blob: single-record operations — rewrite in place (same length)

29.7 ns

blob: single-record operations — offsetOf

4.28 ns

blob: single-record operations — free + rewrite

152 ns

dirty: mark (per-write span coalescing) — contiguous marks

12.8 ns

dirty: mark (per-write span coalescing) — scattered marks

17.8 ns

dirty: take (a frame's drain) — gpu

598 ns

images: registry bookkeeping — acquire (existing url)

205 ns

images: registry bookkeeping — acquire (new url) + release

535 ns

charts: sheet apply, chart vs no chart — no chart

2.22 ms

charts: sheet apply, chart vs no chart — pie from data (blob record per node)

2.69 ms

charts: data write refreshing chart-values — gpu

1.49 ms

1 ns10 ns100 ns1 µs10 µs100 µs1 ms10 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
id index: set x 2000 (build)	gpu	264 µs	275 µs	449 µs	253 µs	2548
id index: single-key operations	get (probe)	65.8 ns	66.8 ns	138 ns	54.5 ns	2547
id index: single-key operations	has	51.6 ns	53.4 ns	84.2 ns	49.6 ns	3204
id index: single-key operations	idAt (memoized hit)	9.02 ns	9.51 ns	18.3 ns	8.71 ns	17824
id index: single-key operations	hashAt	7.01 ns	7.23 ns	14.0 ns	6.79 ns	23430
id index: remove + re-set (churn, drives the round-11 reclaim)	gpu	232 ns	239 ns	357 ns	226 ns	700
adjacency: rebuild (4000 edges, two counting passes)	gpu	39.1 µs	46.4 µs	224 µs	36.8 µs	14940
adjacency: reads	outDegree (O(1))	6.26 ns	6.37 ns	9.47 ns	5.95 ns	26642
adjacency: reads	outEdges	45.2 ns	46.5 ns	69.7 ns	43.5 ns	3678
adjacency: reads	connectedEdges	117 ns	177 ns	2.05 µs	113 ns	947
adjacency: addEdge + removeEdge (the overlay path)	gpu	140 ns	144 ns	192 ns	139 ns	1181
blob: write x 2000 (build)	gpu	104 µs	127 µs	672 µs	100 µs	5491
blob: single-record operations	rewrite in place (same length)	29.7 ns	30.5 ns	44.5 ns	28.7 ns	5607
blob: single-record operations	offsetOf	4.28 ns	4.42 ns	9.87 ns	4.09 ns	38063
blob: single-record operations	free + rewrite	152 ns	172 ns	495 ns	147 ns	991
dirty: mark (per-write span coalescing)	contiguous marks	12.8 ns	13.3 ns	23.5 ns	12.4 ns	12786
dirty: mark (per-write span coalescing)	scattered marks	17.8 ns	18.2 ns	28.1 ns	17.4 ns	9343
dirty: take (a frame's drain)	gpu	598 ns	699 ns	1.27 µs	582 ns	236
images: registry bookkeeping	acquire (existing url)	205 ns	208 ns	253 ns	202 ns	821
images: registry bookkeeping	acquire (new url) + release	535 ns	602 ns	2.27 µs	517 ns	272
charts: sheet apply, chart vs no chart	no chart	2.22 ms	2.28 ms	3.28 ms	2.13 ms	297
charts: sheet apply, chart vs no chart	pie from data (blob record per node)	2.69 ms	2.72 ms	3.11 ms	2.58 ms	252
charts: data write refreshing chart-values	gpu	1.49 ms	1.53 ms	2.55 ms	1.44 ms	453

surface N=2,000 · 8.8 minv4 (gpu) v3

viewport: zoom() get

4.7×

viewport: zoom( n ) set

1.90×

viewport: pan() get

1.09×

viewport: pan( p ) set

1.79×

viewport: panBy()

1.64×

viewport: extent()

1.50×

viewport: renderedExtent()

1.63×

viewport: width() + height() + size() (3 calls)

2.9×

viewport: minZoom/maxZoom get (2 calls)

15×

viewport: center( eles )

18×

viewport: reset()

105×

viewport: viewport({ zoom, pan })

59×

viewport: getCenterPan( eles ) — gpu

6.59 µs

core: instanceString + headless + styleEnabled (3 calls)

1.27×

core: hasElementWithId + hasCompoundNodes (2 calls)

1.19×

core: options()

87×

core: mutableElements() (v4 materializes; v3 is O(1))

1.36×

core: gating getters (5 calls)

1.18×

core: gating setters round-trip (autolock/autoungrabify/autounselectify)

1.90×

core: interaction tuning getters (5 calls) — gpu

7.26 ns

core: graph data() set + get (v3 restyles the graph)

8137×

core: graph scratch() set + get (v3 restyles the graph)

16367×

core: batch( fn ) (empty)

6.3×

core: startBatch + endBatch + batching

5.9×

core: collection()

7.2×

core: compact() (no dead slots — the early-out) — gpu

17.0 µs

collection: element accessors (id/group/isNode/isEdge, 4 calls)

1.76×

collection: length + empty + nonempty (3 calls)

4.6×

collection: eq/first/last (3 calls)

2.3×

collection: slice()

2.2×

collection: toArray()

12×

collection: sort( fn )

1.62×

collection: reduce( fn )

1.40×

collection: min + max ( fn ) (2 calls)

2.0×

collection: some + every (2 calls)

1.62×

collection: same + anySame + contains (3 calls)

75×

collection: allAre (all match: the full walk)

1.55×

collection: is (none match: the full walk)

2.5×

collection: union

6.0×

collection: intersection

29×

collection: difference

29×

collection: symmetricDifference

3.7×

collection: absoluteComplement

15×

collection: byGroup

3.0×

collection: diff

3.3×

collection: indexOf

1.50×

traversal: openNeighborhood + closedNeighborhood (2 calls)

2.9×

traversal: connectedNodes

3.3×

traversal: sources + targets (2 calls)

3.4×

traversal: source() + target() (2 calls)

1.35×

traversal: edgesWith

1.27×

traversal: edgesTo

1.28×

traversal: parallelEdges + codirectedEdges (2 calls)

3.5×

traversal: roots + leaves (2 calls)

13×

traversal: allAreNeighbors

5.3×

traversal: isLoop + isSimple (2 calls)

3.1×

degree: indegree + outdegree (2 calls)

77×

degree: minDegree + maxDegree (2 calls)

44×

degree: minIndegree + maxOutdegree (2 calls)

26×

dimensions: renderedPosition get

2.4×

dimensions: relativePosition get

11×

dimensions: width + height + outerWidth + outerHeight (4 calls)

2.7×

dimensions: rendered dims (4 calls)

3.0×

dimensions: renderedBoundingBox

1.93×

dimensions: midpoint + renderedMidpoint (2 calls) — gpu

1.17 µs

dimensions: source/targetEndpoint (2 calls) — gpu

386 ns

dimensions: rendered source/targetEndpoint (2 calls) — gpu

408 ns

dimensions: shift + silentShift (round-trip)

61×

dimensions: silentPosition

16×

dimensions: positions( fn ) over 100

50×

dimensions: padding + paddedWidth + paddedHeight (3 calls)

13×

curves: isBundledBezier — gpu

25.4 ns

curves: controlPoints + renderedControlPoints (2 calls) — gpu

532 ns

curves: segmentPoints (undefined for bezier) — gpu

41.9 ns

compound: parent + parents + ancestors (3 calls)

3.7×

compound: children + descendants + siblings (3 calls)

4.8×

compound: orphans + nonorphans (2 calls)

6.9×

compound: commonAncestors

124×

compound: isParent/isChild/isChildless/isOrphan (4 calls)

1.56×

flags: select + unselect + selected (band)

15×

flags: selectify + unselectify + selectable

88×

flags: lock + unlock + locked

88×

flags: grabify + ungrabify + grabbable

87×

flags: show + hide + visible

227×

flags: active/activate/unactivate (3 calls)

15×

flags: panify + unpanify + pannable

89×

flags: takesUpSpace + interactive + transparent (3 calls)

3.1×

flags: effectiveOpacity

2.3×

element data: scratch get + set

7.6×

element data: json() one element

2.1×

element data: jsons() over 100

1.99×

element data: label() — gpu

15.2 ns

core: batchData()

8.0×

core: animated()

4.9×

core: zoomRange get — gpu

2.16 ns

core: forceRender() (headless no-op)

14×

core: resize() / invalidateSize()

2.4×

core: onRender + offRender

2.4×

core: removeScratch()

8942×

core: selectionType get

1.99×

core: isReady()

2.2×

core: window()

2.8×

core: container()

2.7×

core: destroyed()

2.2×

core: removeAllListeners()

8.5×

collection: cy()

3.6×

collection: indexOfId()

1.79×

collection: removed() + inside() (2 calls)

1.17×

collection: silentPositions( fn ) over 100

30×

collection: animated()

2.9×

collection: delayAnimation()

2.8×

collection: removeScratch()

9.0×

collection: component()

43×

collection: layoutDimensions()

3.6×

degree: maxIndegree + minOutdegree (2 calls)

25×

flags: grabbed + hidden + inactive (3 calls)

1.99×

curves: renderedSegmentPoints — gpu

36.7 ns

layout contract: edgeSlots + endpoints + degreeOf — gpu

34.6 µs

formats: isColumnarElements + isSerializedElements + isPackedIds — gpu

4.61 ns

0.100 ns1 ns10 ns100 ns1 µs10 µs100 µs1 ms10 ms100 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
viewport: zoom() get	v3	5.63 ns	5.74 ns	9.22 ns	5.45 ns	28768
viewport: zoom() get	gpu	1.21 ns	1.24 ns	1.75 ns	1.21 ns	128095
viewport: zoom( n ) set	v3	612 ns	619 ns	711 ns	590 ns	267
viewport: zoom( n ) set	gpu	321 ns	326 ns	368 ns	308 ns	516
viewport: pan() get	v3	5.43 ns	5.64 ns	11.9 ns	5.17 ns	29859
viewport: pan() get	gpu	4.96 ns	5.22 ns	12.7 ns	4.85 ns	32302
viewport: pan( p ) set	v3	611 ns	645 ns	1.03 µs	598 ns	259
viewport: pan( p ) set	gpu	342 ns	356 ns	596 ns	325 ns	475
viewport: panBy()	v3	566 ns	573 ns	692 ns	554 ns	292
viewport: panBy()	gpu	344 ns	350 ns	693 ns	311 ns	481
viewport: extent()	v3	29.7 ns	45.2 ns	196 ns	28.4 ns	3742
viewport: extent()	gpu	19.8 ns	27.8 ns	114 ns	18.3 ns	6082
viewport: renderedExtent()	v3	27.0 ns	28.7 ns	58.4 ns	25.6 ns	5866
viewport: renderedExtent()	gpu	16.5 ns	23.9 ns	103 ns	15.3 ns	7071
viewport: width() + height() + size() (3 calls)	v3	21.1 ns	21.8 ns	43.2 ns	20.1 ns	7821
viewport: width() + height() + size() (3 calls)	gpu	7.18 ns	7.58 ns	15.7 ns	6.97 ns	22273
viewport: minZoom/maxZoom get (2 calls)	v3	14.8 ns	15.4 ns	25.0 ns	14.4 ns	11066
viewport: minZoom/maxZoom get (2 calls)	gpu	0.999 ns	1.07 ns	2.00 ns	0.997 ns	146337
viewport: center( eles )	v3	117 µs	122 µs	186 µs	112 µs	5706
viewport: center( eles )	gpu	6.57 µs	6.58 µs	6.72 µs	6.44 µs	21
viewport: reset()	v3	938 ns	1.02 µs	1.57 µs	909 ns	161
viewport: reset()	gpu	8.96 ns	9.27 ns	17.6 ns	8.52 ns	18142
viewport: viewport({ zoom, pan })	v3	939 ns	1.02 µs	1.56 µs	910 ns	161
viewport: viewport({ zoom, pan })	gpu	16.0 ns	16.8 ns	38.6 ns	15.2 ns	10118
viewport: getCenterPan( eles )	gpu	6.59 µs	6.64 µs	6.84 µs	6.54 µs	22
core: instanceString + headless + styleEnabled (3 calls)	v3	32.6 ns	34.0 ns	60.0 ns	31.6 ns	5013
core: instanceString + headless + styleEnabled (3 calls)	gpu	25.8 ns	27.0 ns	52.0 ns	24.9 ns	6313
core: hasElementWithId + hasCompoundNodes (2 calls)	v3	39.3 ns	40.2 ns	51.0 ns	38.1 ns	4268
core: hasElementWithId + hasCompoundNodes (2 calls)	gpu	33.0 ns	34.1 ns	46.5 ns	32.4 ns	5005
core: options()	v3	93.4 ns	97.9 ns	167 ns	92.7 ns	1744
core: options()	gpu	1.07 ns	1.18 ns	2.20 ns	1.05 ns	135407
core: mutableElements() (v4 materializes; v3 is O(1))	v3	3.19 ns	3.32 ns	6.59 ns	3.14 ns	50247
core: mutableElements() (v4 materializes; v3 is O(1))	gpu	2.35 ns	2.40 ns	5.52 ns	2.25 ns	68972
core: gating getters (5 calls)	v3	15.7 ns	16.1 ns	24.1 ns	15.0 ns	10645
core: gating getters (5 calls)	gpu	13.3 ns	13.8 ns	24.6 ns	12.5 ns	12416
core: gating setters round-trip (autolock/autoungrabify/autounselectify)	v3	11.4 ns	12.2 ns	21.6 ns	10.8 ns	13924
core: gating setters round-trip (autolock/autoungrabify/autounselectify)	gpu	6.03 ns	6.37 ns	14.3 ns	5.79 ns	26630
core: interaction tuning getters (5 calls)	gpu	7.26 ns	7.69 ns	16.3 ns	7.15 ns	22081
core: graph data() set + get (v3 restyles the graph)	v3	2.68 ms	2.71 ms	3.76 ms	2.50 ms	255
core: graph data() set + get (v3 restyles the graph)	gpu	329 ns	336 ns	424 ns	321 ns	504
core: graph scratch() set + get (v3 restyles the graph)	v3	2.71 ms	2.71 ms	3.63 ms	2.48 ms	254
core: graph scratch() set + get (v3 restyles the graph)	gpu	165 ns	170 ns	255 ns	164 ns	1003
core: batch( fn ) (empty)	v3	533 ns	641 ns	1.64 µs	510 ns	262
core: batch( fn ) (empty)	gpu	84.9 ns	87.8 ns	123 ns	83.3 ns	1926
core: startBatch + endBatch + batching	v3	530 ns	644 ns	1.71 µs	496 ns	261
core: startBatch + endBatch + batching	gpu	89.6 ns	164 ns	3.00 µs	86.4 ns	1005
core: collection()	v3	448 ns	531 ns	1.56 µs	439 ns	316
core: collection()	gpu	62.0 ns	65.8 ns	150 ns	59.5 ns	2580
core: compact() (no dead slots — the early-out)	gpu	17.0 µs	21.5 µs	95.3 µs	16.4 µs	31800
collection: element accessors (id/group/isNode/isEdge, 4 calls)	v3	60.8 ns	65.8 ns	200 ns	57.3 ns	2584
collection: element accessors (id/group/isNode/isEdge, 4 calls)	gpu	34.5 ns	36.4 ns	73.4 ns	33.7 ns	4647
collection: length + empty + nonempty (3 calls)	v3	6.72 ns	6.92 ns	12.5 ns	6.34 ns	24549
collection: length + empty + nonempty (3 calls)	gpu	1.47 ns	1.52 ns	1.99 ns	1.44 ns	106292
collection: eq/first/last (3 calls)	v3	15.2 ns	15.6 ns	24.7 ns	14.9 ns	10896
collection: eq/first/last (3 calls)	gpu	6.64 ns	6.91 ns	14.1 ns	6.33 ns	24560
collection: slice()	v3	3.73 µs	3.82 µs	4.23 µs	3.59 µs	41
collection: slice()	gpu	1.72 µs	1.73 µs	1.89 µs	1.67 µs	94
collection: toArray()	v3	8.85 µs	8.97 µs	9.29 µs	8.69 µs	15
collection: toArray()	gpu	757 ns	770 ns	933 ns	718 ns	212
collection: sort( fn )	v3	14.7 µs	15.1 µs	15.4 µs	14.1 µs	12
collection: sort( fn )	gpu	9.10 µs	9.14 µs	9.36 µs	8.99 µs	15
collection: reduce( fn )	v3	157 ns	160 ns	188 ns	117 ns	1062
collection: reduce( fn )	gpu	112 ns	113 ns	130 ns	111 ns	1509
collection: min + max ( fn ) (2 calls)	v3	11.3 µs	11.3 µs	11.6 µs	11.0 µs	11
collection: min + max ( fn ) (2 calls)	gpu	5.60 µs	5.63 µs	5.92 µs	5.50 µs	26
collection: some + every (2 calls)	v3	894 ns	908 ns	1.04 µs	883 ns	183
collection: some + every (2 calls)	gpu	551 ns	559 ns	595 ns	549 ns	304
collection: same + anySame + contains (3 calls)	v3	7.58 µs	7.71 µs	8.26 µs	7.48 µs	18
collection: same + anySame + contains (3 calls)	gpu	101 ns	105 ns	178 ns	98.2 ns	1627
collection: allAre (all match: the full walk)	v3	3.64 µs	3.65 µs	4.02 µs	3.41 µs	42
collection: allAre (all match: the full walk)	gpu	2.34 µs	2.40 µs	2.84 µs	2.34 µs	64
collection: is (none match: the full walk)	v3	5.67 µs	5.71 µs	6.01 µs	5.42 µs	26
collection: is (none match: the full walk)	gpu	2.27 µs	2.32 µs	3.38 µs	2.22 µs	68
collection: union	v3	98.3 µs	108 µs	158 µs	94.4 µs	6405
collection: union	gpu	16.4 µs	16.6 µs	16.8 µs	15.7 µs	12
collection: intersection	v3	93.5 µs	104 µs	180 µs	90.6 µs	6638
collection: intersection	gpu	3.21 µs	3.24 µs	3.47 µs	3.16 µs	48
collection: difference	v3	93.0 µs	105 µs	212 µs	91.6 µs	6610
collection: difference	gpu	3.16 µs	3.18 µs	3.49 µs	3.10 µs	49
collection: symmetricDifference	v3	24.6 µs	25.0 µs	25.0 µs	24.4 µs	12
collection: symmetricDifference	gpu	6.58 µs	6.62 µs	6.81 µs	6.48 µs	22
collection: absoluteComplement	v3	6.23 ms	6.74 ms	10.5 ms	6.04 ms	100
collection: absoluteComplement	gpu	410 µs	419 µs	549 µs	404 µs	1670
collection: byGroup	v3	995 µs	1.01 ms	1.25 ms	984 µs	692
collection: byGroup	gpu	328 µs	341 µs	615 µs	314 µs	2054
collection: diff	v3	37.3 µs	37.5 µs	38.8 µs	35.1 µs	12
collection: diff	gpu	11.3 µs	11.4 µs	12.0 µs	9.84 µs	11
collection: indexOf	v3	34.9 ns	36.4 ns	75.9 ns	34.0 ns	4696
collection: indexOf	gpu	23.4 ns	24.3 ns	42.8 ns	22.5 ns	7044
traversal: openNeighborhood + closedNeighborhood (2 calls)	v3	9.54 µs	9.77 µs	10.5 µs	9.39 µs	14
traversal: openNeighborhood + closedNeighborhood (2 calls)	gpu	3.25 µs	3.43 µs	4.01 µs	3.08 µs	45
traversal: connectedNodes	v3	1.02 µs	1.11 µs	1.88 µs	983 ns	149
traversal: connectedNodes	gpu	311 ns	320 ns	479 ns	304 ns	531
traversal: sources + targets (2 calls)	v3	1.77 µs	1.96 µs	2.73 µs	1.71 µs	84
traversal: sources + targets (2 calls)	gpu	528 ns	734 ns	2.40 µs	514 ns	226
traversal: source() + target() (2 calls)	v3	77.4 ns	79.4 ns	118 ns	64.5 ns	2154
traversal: source() + target() (2 calls)	gpu	57.3 ns	59.4 ns	99.2 ns	54.0 ns	2877
traversal: edgesWith	v3	68.3 µs	70.5 µs	118 µs	67.0 µs	9881
traversal: edgesWith	gpu	53.5 µs	54.0 µs	56.8 µs	52.7 µs	12
traversal: edgesTo	v3	68.2 µs	70.3 µs	118 µs	66.8 µs	9894
traversal: edgesTo	gpu	53.2 µs	53.8 µs	55.2 µs	52.7 µs	12
traversal: parallelEdges + codirectedEdges (2 calls)	v3	3.75 µs	3.98 µs	4.77 µs	3.57 µs	39
traversal: parallelEdges + codirectedEdges (2 calls)	gpu	1.08 µs	1.30 µs	2.15 µs	1.05 µs	124
traversal: roots + leaves (2 calls)	v3	5.05 ms	5.47 ms	9.04 ms	4.87 ms	124
traversal: roots + leaves (2 calls)	gpu	387 µs	396 µs	584 µs	379 µs	1768
traversal: allAreNeighbors	v3	198 µs	213 µs	363 µs	193 µs	3270
traversal: allAreNeighbors	gpu	37.3 µs	37.7 µs	38.3 µs	37.0 µs	12
traversal: isLoop + isSimple (2 calls)	v3	156 ns	158 ns	178 ns	145 ns	1062
traversal: isLoop + isSimple (2 calls)	gpu	50.3 ns	51.8 ns	75.3 ns	49.3 ns	3299
degree: indegree + outdegree (2 calls)	v3	3.59 µs	3.95 µs	4.77 µs	3.46 µs	39
degree: indegree + outdegree (2 calls)	gpu	46.8 ns	48.2 ns	76.2 ns	46.0 ns	3544
degree: minDegree + maxDegree (2 calls)	v3	16.3 ms	17.6 ms	21.6 ms	15.8 ms	35
degree: minDegree + maxDegree (2 calls)	gpu	370 µs	376 µs	505 µs	363 µs	1851
degree: minIndegree + maxOutdegree (2 calls)	v3	9.22 ms	10.0 ms	14.0 ms	8.99 ms	66
degree: minIndegree + maxOutdegree (2 calls)	gpu	359 µs	368 µs	594 µs	348 µs	1910
dimensions: renderedPosition get	v3	73.7 ns	77.4 ns	144 ns	72.1 ns	2196
dimensions: renderedPosition get	gpu	30.5 ns	33.9 ns	108 ns	29.0 ns	4976
dimensions: relativePosition get	v3	860 ns	880 ns	1.15 µs	836 ns	189
dimensions: relativePosition get	gpu	76.1 ns	81.7 ns	192 ns	74.1 ns	2073
dimensions: width + height + outerWidth + outerHeight (4 calls)	v3	618 ns	624 ns	666 ns	613 ns	273
dimensions: width + height + outerWidth + outerHeight (4 calls)	gpu	232 ns	234 ns	260 ns	226 ns	726
dimensions: rendered dims (4 calls)	v3	717 ns	721 ns	782 ns	706 ns	235
dimensions: rendered dims (4 calls)	gpu	238 ns	241 ns	260 ns	238 ns	710
dimensions: renderedBoundingBox	v3	1.79 µs	1.81 µs	2.16 µs	1.74 µs	89
dimensions: renderedBoundingBox	gpu	926 ns	942 ns	1.12 µs	900 ns	175
dimensions: midpoint + renderedMidpoint (2 calls)	gpu	1.17 µs	1.18 µs	1.30 µs	1.13 µs	139
dimensions: source/targetEndpoint (2 calls)	gpu	386 ns	393 ns	443 ns	382 ns	431
dimensions: rendered source/targetEndpoint (2 calls)	gpu	408 ns	414 ns	511 ns	398 ns	410
dimensions: shift + silentShift (round-trip)	v3	292 µs	304 µs	570 µs	285 µs	2301
dimensions: shift + silentShift (round-trip)	gpu	4.82 µs	4.84 µs	5.02 µs	4.74 µs	31
dimensions: silentPosition	v3	903 ns	917 ns	1.14 µs	861 ns	183
dimensions: silentPosition	gpu	55.8 ns	58.3 ns	83.8 ns	54.8 ns	2915
dimensions: positions( fn ) over 100	v3	178 µs	183 µs	250 µs	173 µs	3816
dimensions: positions( fn ) over 100	gpu	3.54 µs	3.57 µs	3.81 µs	3.47 µs	44
dimensions: padding + paddedWidth + paddedHeight (3 calls)	v3	2.03 µs	2.04 µs	2.21 µs	1.98 µs	80
dimensions: padding + paddedWidth + paddedHeight (3 calls)	gpu	161 ns	163 ns	201 ns	155 ns	1052
curves: isBundledBezier	gpu	25.4 ns	26.0 ns	35.7 ns	21.2 ns	6538
curves: controlPoints + renderedControlPoints (2 calls)	gpu	532 ns	547 ns	804 ns	526 ns	300
curves: segmentPoints (undefined for bezier)	gpu	41.9 ns	43.7 ns	81.3 ns	36.0 ns	3910
compound: parent + parents + ancestors (3 calls)	v3	3.06 µs	3.34 µs	4.17 µs	2.94 µs	47
compound: parent + parents + ancestors (3 calls)	gpu	817 ns	1.15 µs	2.42 µs	793 ns	140
compound: children + descendants + siblings (3 calls)	v3	8.79 µs	8.86 µs	9.80 µs	7.89 µs	16
compound: children + descendants + siblings (3 calls)	gpu	1.83 µs	2.37 µs	3.87 µs	1.78 µs	66
compound: orphans + nonorphans (2 calls)	v3	3.71 µs	4.01 µs	4.82 µs	3.58 µs	38
compound: orphans + nonorphans (2 calls)	gpu	539 ns	783 ns	3.18 µs	531 ns	212
compound: commonAncestors	v3	9.59 µs	9.67 µs	10.2 µs	9.25 µs	14
compound: commonAncestors	gpu	77.1 ns	80.2 ns	132 ns	73.9 ns	2121
compound: isParent/isChild/isChildless/isOrphan (4 calls)	v3	70.6 ns	73.5 ns	118 ns	69.7 ns	2336
compound: isParent/isChild/isChildless/isOrphan (4 calls)	gpu	45.3 ns	47.2 ns	76.1 ns	43.2 ns	3616
flags: select + unselect + selected (band)	v3	181 µs	190 µs	391 µs	178 µs	3686
flags: select + unselect + selected (band)	gpu	12.1 µs	12.2 µs	12.3 µs	11.9 µs	10
flags: selectify + unselectify + selectable	v3	181 µs	186 µs	316 µs	176 µs	3758
flags: selectify + unselectify + selectable	gpu	2.05 µs	2.08 µs	2.45 µs	2.00 µs	79
flags: lock + unlock + locked	v3	179 µs	185 µs	294 µs	174 µs	3785
flags: lock + unlock + locked	gpu	2.04 µs	2.05 µs	2.24 µs	2.01 µs	80
flags: grabify + ungrabify + grabbable	v3	178 µs	184 µs	295 µs	175 µs	3805
flags: grabify + ungrabify + grabbable	gpu	2.04 µs	2.04 µs	2.14 µs	1.99 µs	80
flags: show + hide + visible	v3	2.57 ms	2.71 ms	4.03 ms	2.49 ms	253
flags: show + hide + visible	gpu	11.3 µs	11.3 µs	11.4 µs	11.2 µs	11
flags: active/activate/unactivate (3 calls)	v3	180 µs	185 µs	286 µs	176 µs	3772
flags: active/activate/unactivate (3 calls)	gpu	11.9 µs	11.9 µs	12.0 µs	11.8 µs	11
flags: panify + unpanify + pannable	v3	182 µs	188 µs	345 µs	176 µs	3706
flags: panify + unpanify + pannable	gpu	2.03 µs	2.04 µs	2.22 µs	1.99 µs	81
flags: takesUpSpace + interactive + transparent (3 calls)	v3	195 ns	201 ns	298 ns	194 ns	852
flags: takesUpSpace + interactive + transparent (3 calls)	gpu	63.3 ns	65.6 ns	90.5 ns	62.0 ns	2606
flags: effectiveOpacity	v3	49.1 ns	50.1 ns	60.7 ns	46.9 ns	3429
flags: effectiveOpacity	gpu	21.3 ns	21.6 ns	31.3 ns	20.6 ns	7947
element data: scratch get + set	v3	1.30 µs	1.32 µs	1.59 µs	1.26 µs	125
element data: scratch get + set	gpu	171 ns	177 ns	267 ns	166 ns	966
element data: json() one element	v3	300 ns	389 ns	3.46 µs	292 ns	421
element data: json() one element	gpu	141 ns	183 ns	542 ns	137 ns	875
element data: jsons() over 100	v3	39.8 µs	40.4 µs	42.6 µs	37.2 µs	12
element data: jsons() over 100	gpu	20.0 µs	21.0 µs	24.0 µs	16.3 µs	12
element data: label()	gpu	15.2 ns	15.9 ns	27.1 ns	14.5 ns	10799
core: batchData()	v3	2.29 µs	2.41 µs	3.06 µs	2.25 µs	66
core: batchData()	gpu	285 ns	380 ns	2.26 µs	279 ns	432
core: animated()	v3	7.53 ns	7.81 ns	13.9 ns	7.37 ns	21723
core: animated()	gpu	1.53 ns	1.60 ns	2.35 ns	1.50 ns	103231
core: zoomRange get	gpu	2.16 ns	2.28 ns	6.12 ns	2.05 ns	73373
core: forceRender() (headless no-op)	v3	20.1 ns	20.8 ns	32.9 ns	19.6 ns	8235
core: forceRender() (headless no-op)	gpu	1.48 ns	1.54 ns	1.94 ns	1.48 ns	106961
core: resize() / invalidateSize()	v3	304 ns	308 ns	378 ns	297 ns	546
core: resize() / invalidateSize()	gpu	127 ns	131 ns	163 ns	126 ns	1300
core: onRender + offRender	v3	472 ns	549 ns	1.26 µs	457 ns	306
core: onRender + offRender	gpu	194 ns	240 ns	758 ns	189 ns	696
core: removeScratch()	v3	2.50 ms	2.58 ms	4.14 ms	2.39 ms	266
core: removeScratch()	gpu	279 ns	285 ns	387 ns	268 ns	596
core: selectionType get	v3	4.51 ns	4.79 ns	9.74 ns	4.35 ns	35404
core: selectionType get	gpu	2.26 ns	2.46 ns	6.18 ns	2.10 ns	68002
core: isReady()	v3	3.25 ns	3.37 ns	4.56 ns	3.25 ns	49944
core: isReady()	gpu	1.48 ns	1.53 ns	1.87 ns	1.48 ns	107680
core: window()	v3	4.11 ns	4.28 ns	7.79 ns	4.10 ns	39371
core: window()	gpu	1.48 ns	1.54 ns	1.98 ns	1.45 ns	107056
core: container()	v3	3.94 ns	4.08 ns	7.82 ns	3.75 ns	41306
core: container()	gpu	1.48 ns	1.53 ns	1.90 ns	1.48 ns	107366
core: destroyed()	v3	3.30 ns	3.42 ns	6.65 ns	3.25 ns	49378
core: destroyed()	gpu	1.48 ns	1.54 ns	1.94 ns	1.45 ns	106918
core: removeAllListeners()	v3	13.5 ns	14.2 ns	26.2 ns	13.0 ns	11980
core: removeAllListeners()	gpu	1.58 ns	1.64 ns	1.98 ns	1.58 ns	101034
collection: cy()	v3	7.80 ns	8.39 ns	15.6 ns	7.32 ns	20373
collection: cy()	gpu	2.16 ns	2.44 ns	6.29 ns	2.06 ns	68322
collection: indexOfId()	v3	69.6 ns	71.8 ns	123 ns	67.6 ns	2381
collection: indexOfId()	gpu	38.8 ns	40.2 ns	71.9 ns	38.1 ns	4252
collection: removed() + inside() (2 calls)	v3	17.7 ns	18.4 ns	30.0 ns	17.3 ns	9326
collection: removed() + inside() (2 calls)	gpu	15.1 ns	15.7 ns	25.4 ns	14.6 ns	10938
collection: silentPositions( fn ) over 100	v3	107 µs	115 µs	266 µs	105 µs	6055
collection: silentPositions( fn ) over 100	gpu	3.53 µs	3.56 µs	3.83 µs	3.47 µs	44
collection: animated()	v3	12.2 ns	12.5 ns	22.6 ns	11.9 ns	13651
collection: animated()	gpu	4.19 ns	4.34 ns	7.51 ns	4.08 ns	38925
collection: delayAnimation()	v3	456 ns	594 ns	2.00 µs	444 ns	281
collection: delayAnimation()	gpu	165 ns	175 ns	309 ns	159 ns	965
collection: removeScratch()	v3	1.81 µs	1.83 µs	2.32 µs	1.77 µs	89
collection: removeScratch()	gpu	202 ns	209 ns	287 ns	200 ns	804
collection: component()	v3	79.7 ms	82.1 ms	90.1 ms	75.0 ms	12
collection: component()	gpu	1.84 ms	2.09 ms	5.46 ms	1.78 ms	329
collection: layoutDimensions()	v3	618 ns	625 ns	805 ns	600 ns	270
collection: layoutDimensions()	gpu	173 ns	181 ns	322 ns	168 ns	940
degree: maxIndegree + minOutdegree (2 calls)	v3	8.87 ms	9.57 ms	13.6 ms	8.66 ms	69
degree: maxIndegree + minOutdegree (2 calls)	gpu	353 µs	367 µs	588 µs	347 µs	1911
flags: grabbed + hidden + inactive (3 calls)	v3	85.9 ns	88.2 ns	137 ns	81.2 ns	1946
flags: grabbed + hidden + inactive (3 calls)	gpu	43.2 ns	44.5 ns	58.5 ns	40.3 ns	3842
curves: renderedSegmentPoints	gpu	36.7 ns	37.8 ns	73.9 ns	35.6 ns	4526
layout contract: edgeSlots + endpoints + degreeOf	gpu	34.6 µs	36.7 µs	66.4 µs	29.4 µs	18774
formats: isColumnarElements + isSerializedElements + isPackedIds	gpu	4.61 ns	4.84 ns	9.04 ns	4.60 ns	34855

compaction N=20,000 · 53.8 s

cpu pick: background miss (full highWater walk) — peak

223 µs

cpu pick: background miss (full highWater walk) — compacted

50.2 µs

cpu pick: node hit — peak

223 µs

cpu pick: node hit — compacted

50.2 µs

isCurrent: current ref (forwards present vs absent) — no forwards (peak)

5.88 ns

isCurrent: current ref (forwards present vs absent) — forwards present (compacted)

5.98 ns

isCurrent: stale-ref repair (chase + rewrite, alloc included) — current (baseline alloc)

8.90 ns

isCurrent: stale-ref repair (chase + rewrite, alloc included) — stale → repair

50.3 ns

control: nodes({ selected }) scan (order-list, expected parity) — peak

23.7 µs

control: nodes({ selected }) scan (order-list, expected parity) — compacted

22.8 µs

control: whole-graph boundingBox (expected parity) — peak

104 µs

control: whole-graph boundingBox (expected parity) — compacted

100 µs

1 ns10 ns100 ns1 µs10 µs100 µs1 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
cpu pick: background miss (full highWater walk)	peak	223 µs	228 µs	322 µs	216 µs	3076
cpu pick: background miss (full highWater walk)	compacted	50.2 µs	50.5 µs	51.4 µs	49.4 µs	12
cpu pick: node hit	peak	223 µs	228 µs	277 µs	221 µs	3084
cpu pick: node hit	compacted	50.2 µs	51.3 µs	53.9 µs	49.2 µs	12
isCurrent: current ref (forwards present vs absent)	no forwards (peak)	5.88 ns	6.07 ns	10.6 ns	5.62 ns	27925
isCurrent: current ref (forwards present vs absent)	forwards present (compacted)	5.98 ns	6.25 ns	13.2 ns	5.63 ns	27124
isCurrent: stale-ref repair (chase + rewrite, alloc included)	current (baseline alloc)	8.90 ns	9.70 ns	22.1 ns	8.48 ns	17367
isCurrent: stale-ref repair (chase + rewrite, alloc included)	stale → repair	50.3 ns	56.0 ns	108 ns	47.7 ns	3053
control: nodes({ selected }) scan (order-list, expected parity)	peak	23.7 µs	26.5 µs	33.6 µs	21.6 µs	12
control: nodes({ selected }) scan (order-list, expected parity)	compacted	22.8 µs	23.9 µs	25.2 µs	22.3 µs	12
control: whole-graph boundingBox (expected parity)	peak	104 µs	108 µs	171 µs	101 µs	6423
control: whole-graph boundingBox (expected parity)	compacted	100 µs	102 µs	149 µs	96.4 µs	6752

compound N=2,000 · 27.8 sv4 (gpu) v3

parent drag (shift subtree + bb settle)

185×

child drag (move + parent re-derive)

51×

reparent round-trip

76×

1 µs10 µs100 µs1 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
parent drag (shift subtree + bb settle)	v3	260 µs	291 µs	696 µs	246 µs	2387
parent drag (shift subtree + bb settle)	gpu	1.40 µs	1.42 µs	1.61 µs	1.36 µs	112
child drag (move + parent re-derive)	v3	81.7 µs	89.6 µs	151 µs	77.5 µs	7623
child drag (move + parent re-derive)	gpu	1.59 µs	1.61 µs	1.82 µs	1.53 µs	101
reparent round-trip	v3	84.6 µs	93.3 µs	245 µs	81.6 µs	7373
reparent round-trip	gpu	1.11 µs	1.13 µs	1.40 µs	1.07 µs	142

curves N=2,000 · 1.7 s

curve premium: build (style apply + derivation) — curved

10.3 ms

curve premium: build (style apply + derivation) — straight

8.90 ms

curve premium: node.position x 1000 (drag) — curved

376 µs

curve premium: node.position x 1000 (drag) — straight

313 µs

curve premium: nodes().positions() (bulk move) — curved

302 µs

curve premium: nodes().positions() (bulk move) — straight

297 µs

curve premium: first read after a bulk move (cold) — curved

5.86 ms

curve premium: first read after a bulk move (cold) — straight

2.73 ms

curve premium: same read again (warm) — curved

2.21 ms

curve premium: same read again (warm) — straight

3.20 ms

curve premium: edge.midpoint x 4000 — curved

1.83 ms

curve premium: edge.midpoint x 4000 — straight

3.11 ms

curve premium: edge.controlPoints x 4000 — curved

1.95 ms

curve premium: edge.controlPoints x 4000 — straight

1.09 ms

curve premium: cy.elements().boundingBox() x 20 — curved

38.0 ms

curve premium: cy.elements().boundingBox() x 20 — straight

21.2 ms

curve premium: cy.fit() x 20 (store scan) — curved

3.84 ms

curve premium: cy.fit() x 20 (store scan) — straight

3.53 ms

curve premium: elementsInBox x 20 (half the graph) — curved

24.9 ms

curve premium: elementsInBox x 20 (half the graph) — straight

6.10 ms

curve premium: edge.hide/show x 1000 (bundle re-fan) — curved

3.88 ms

curve premium: edge.hide/show x 1000 (bundle re-fan) — straight

2.15 ms

100 µs1 ms10 ms100 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
curve premium: build (style apply + derivation)	curved	10.3 ms	10.3 ms	10.3 ms	10.3 ms	1
curve premium: build (style apply + derivation)	straight	8.90 ms	8.90 ms	8.90 ms	8.90 ms	1
curve premium: node.position x 1000 (drag)	curved	376 µs	376 µs	376 µs	376 µs	1
curve premium: node.position x 1000 (drag)	straight	313 µs	313 µs	313 µs	313 µs	1
curve premium: nodes().positions() (bulk move)	curved	302 µs	302 µs	302 µs	302 µs	1
curve premium: nodes().positions() (bulk move)	straight	297 µs	297 µs	297 µs	297 µs	1
curve premium: first read after a bulk move (cold)	curved	5.86 ms	5.86 ms	5.86 ms	5.86 ms	1
curve premium: first read after a bulk move (cold)	straight	2.73 ms	2.73 ms	2.73 ms	2.73 ms	1
curve premium: same read again (warm)	curved	2.21 ms	2.21 ms	2.21 ms	2.21 ms	1
curve premium: same read again (warm)	straight	3.20 ms	3.20 ms	3.20 ms	3.20 ms	1
curve premium: edge.midpoint x 4000	curved	1.83 ms	1.83 ms	1.83 ms	1.83 ms	1
curve premium: edge.midpoint x 4000	straight	3.11 ms	3.11 ms	3.11 ms	3.11 ms	1
curve premium: edge.controlPoints x 4000	curved	1.95 ms	1.95 ms	1.95 ms	1.95 ms	1
curve premium: edge.controlPoints x 4000	straight	1.09 ms	1.09 ms	1.09 ms	1.09 ms	1
curve premium: cy.elements().boundingBox() x 20	curved	38.0 ms	38.0 ms	38.0 ms	38.0 ms	1
curve premium: cy.elements().boundingBox() x 20	straight	21.2 ms	21.2 ms	21.2 ms	21.2 ms	1
curve premium: cy.fit() x 20 (store scan)	curved	3.84 ms	3.84 ms	3.84 ms	3.84 ms	1
curve premium: cy.fit() x 20 (store scan)	straight	3.53 ms	3.53 ms	3.53 ms	3.53 ms	1
curve premium: elementsInBox x 20 (half the graph)	curved	24.9 ms	24.9 ms	24.9 ms	24.9 ms	1
curve premium: elementsInBox x 20 (half the graph)	straight	6.10 ms	6.10 ms	6.10 ms	6.10 ms	1
curve premium: edge.hide/show x 1000 (bundle re-fan)	curved	3.88 ms	3.88 ms	3.88 ms	3.88 ms	1
curve premium: edge.hide/show x 1000 (bundle re-fan)	straight	2.15 ms	2.15 ms	2.15 ms	2.15 ms	1

arrows N=2,000 · 8.8 s

arrow premium: build (style apply + arrow fold) — arrowed

5.91 ms

arrow premium: build (style apply + arrow fold) — none

5.95 ms

arrow premium: cy.style() re-apply x 4 — arrowed

8.08 ms

arrow premium: cy.style() re-apply x 4 — none

8.05 ms

arrow premium: node.position x 200000 (drag) — arrowed

22.1 ms

arrow premium: node.position x 200000 (drag) — none

22.2 ms

arrow premium: edge.sourceEndpoint x 200000 — arrowed

61.1 ms

arrow premium: edge.sourceEndpoint x 200000 — none

60.8 ms

arrow premium: edge.targetEndpoint x 200000 — arrowed

61.2 ms

arrow premium: edge.targetEndpoint x 200000 — none

60.0 ms

arrow premium: edge.midpoint x 200000 — arrowed

145 ms

arrow premium: edge.midpoint x 200000 — none

144 ms

arrow premium: edge.style('target-arrow-shape') x 200000 — arrowed

32.8 ms

arrow premium: edge.style('target-arrow-shape') x 200000 — none

27.8 ms

arrow premium: cy.elements().boundingBox() x 20 — arrowed

20.3 ms

arrow premium: cy.elements().boundingBox() x 20 — none

20.3 ms

arrow premium: elementsInBox x 20 (half the graph) — arrowed

6.54 ms

arrow premium: elementsInBox x 20 (half the graph) — none

5.71 ms

arrow premium: control: sourceEndpoint, bezier vs straight — arrowed

77.2 ms

arrow premium: control: sourceEndpoint, bezier vs straight — none

60.9 ms

arrow premium: edge column bytes — B/edge

156 ms

1 ms10 ms100 ms1 s

All stats (table view)

group	bench	p50	avg	p99	min	samples
arrow premium: build (style apply + arrow fold)	arrowed	5.91 ms	5.91 ms	5.91 ms	5.91 ms	1
arrow premium: build (style apply + arrow fold)	none	5.95 ms	5.95 ms	5.95 ms	5.95 ms	1
arrow premium: cy.style() re-apply x 4	arrowed	8.08 ms	8.08 ms	8.08 ms	8.08 ms	1
arrow premium: cy.style() re-apply x 4	none	8.05 ms	8.05 ms	8.05 ms	8.05 ms	1
arrow premium: node.position x 200000 (drag)	arrowed	22.1 ms	22.1 ms	22.1 ms	22.1 ms	1
arrow premium: node.position x 200000 (drag)	none	22.2 ms	22.2 ms	22.2 ms	22.2 ms	1
arrow premium: edge.sourceEndpoint x 200000	arrowed	61.1 ms	61.1 ms	61.1 ms	61.1 ms	1
arrow premium: edge.sourceEndpoint x 200000	none	60.8 ms	60.8 ms	60.8 ms	60.8 ms	1
arrow premium: edge.targetEndpoint x 200000	arrowed	61.2 ms	61.2 ms	61.2 ms	61.2 ms	1
arrow premium: edge.targetEndpoint x 200000	none	60.0 ms	60.0 ms	60.0 ms	60.0 ms	1
arrow premium: edge.midpoint x 200000	arrowed	145 ms	145 ms	145 ms	145 ms	1
arrow premium: edge.midpoint x 200000	none	144 ms	144 ms	144 ms	144 ms	1
arrow premium: edge.style('target-arrow-shape') x 200000	arrowed	32.8 ms	32.8 ms	32.8 ms	32.8 ms	1
arrow premium: edge.style('target-arrow-shape') x 200000	none	27.8 ms	27.8 ms	27.8 ms	27.8 ms	1
arrow premium: cy.elements().boundingBox() x 20	arrowed	20.3 ms	20.3 ms	20.3 ms	20.3 ms	1
arrow premium: cy.elements().boundingBox() x 20	none	20.3 ms	20.3 ms	20.3 ms	20.3 ms	1
arrow premium: elementsInBox x 20 (half the graph)	arrowed	6.54 ms	6.54 ms	6.54 ms	6.54 ms	1
arrow premium: elementsInBox x 20 (half the graph)	none	5.71 ms	5.71 ms	5.71 ms	5.71 ms	1
arrow premium: control: sourceEndpoint, bezier vs straight	arrowed	77.2 ms	77.2 ms	77.2 ms	77.2 ms	1
arrow premium: control: sourceEndpoint, bezier vs straight	none	60.9 ms	60.9 ms	60.9 ms	60.9 ms	1
arrow premium: edge column bytes	B/edge	156 ms	156 ms	156 ms	156 ms	1

labels N=20,000 · 1.6 s

labels: breakLines x 20000 — gpu

79.6 ms

labels: estimateBlock x 20000 — gpu

90.7 ms

labels: addNode x 20000 — gpu

30.5 ms

labels: setLabel (build) x 20000 — gpu

104 ms

labels: setLabel (rewrite) x 20000 — gpu

104 ms

labels: boundingBox with 20000 label terms — gpu

3.43 ms

1 ms10 ms100 ms1 s

All stats (table view)

group	bench	p50	avg	p99	min	samples
labels: breakLines x 20000	gpu	79.6 ms	79.6 ms	79.6 ms	79.6 ms	1
labels: estimateBlock x 20000	gpu	90.7 ms	90.7 ms	90.7 ms	90.7 ms	1
labels: addNode x 20000	gpu	30.5 ms	30.5 ms	30.5 ms	30.5 ms	1
labels: setLabel (build) x 20000	gpu	104 ms	104 ms	104 ms	104 ms	1
labels: setLabel (rewrite) x 20000	gpu	104 ms	104 ms	104 ms	104 ms	1
labels: boundingBox with 20000 label terms	gpu	3.43 ms	3.43 ms	3.43 ms	3.43 ms	1

transitions N=2,000 · 28.9 s

transition: auto-extent shift, whole-channel re-derive (2000 slots) — transitions off

1.15 ms

transition: auto-extent shift, whole-channel re-derive (2000 slots) — transitions on

3.65 ms

transition: bulk tween tick (2000 slots, one manager tick) — tick

122 µs

transition: explicit-domain write (O(changed) — one element) — transitions off

14.8 µs

transition: explicit-domain write (O(changed) — one element) — transitions on

19.0 µs

transition: sheet swap, whole-graph apply (2000 slots diff + tween) — transitions off

2.11 ms

transition: sheet swap, whole-graph apply (2000 slots diff + tween) — transitions on

4.85 ms

10 µs100 µs1 ms10 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
transition: auto-extent shift, whole-channel re-derive (2000 slots)	transitions off	1.15 ms	1.20 ms	2.09 ms	1.09 ms	576
transition: auto-extent shift, whole-channel re-derive (2000 slots)	transitions on	3.65 ms	3.85 ms	5.68 ms	3.38 ms	177
transition: bulk tween tick (2000 slots, one manager tick)	tick	122 µs	127 µs	237 µs	117 µs	5391
transition: explicit-domain write (O(changed) — one element)	transitions off	14.8 µs	15.7 µs	18.5 µs	14.4 µs	12
transition: explicit-domain write (O(changed) — one element)	transitions on	19.0 µs	19.4 µs	19.8 µs	18.4 µs	12
transition: sheet swap, whole-graph apply (2000 slots diff + tween)	transitions off	2.11 ms	2.15 ms	2.56 ms	2.05 ms	324
transition: sheet swap, whole-graph apply (2000 slots diff + tween)	transitions on	4.85 ms	4.60 ms	6.69 ms	3.52 ms	148

geometry-tween N=2,000 · 32.6 s

size tick: unlabelled (2000 slots) — tick

570 µs

size tick: labelled, center/center (re-anchor early-out) (2000 slots) — tick

605 µs

size tick: labelled, hanging (sidecar rewrite per tick) (2000 slots) — tick

1.25 ms

edge width tick (4000 slots) — bare

277 µs

edge width tick (4000 slots) — with rides (casing + match-line arrows)

690 µs

padding tick + auto-bounds flush (250 parents × 8 children) — tick + flush

176 µs

font-size tick (2000 labels) — wrap none (dims scale-patch)

368 µs

font-size tick (2000 labels) — wrapped (re-break per tick)

5.69 ms

baseline: paint (opacity) tick (2000 slots) — tick

141 µs

100 µs1 ms10 ms

All stats (table view)

group	bench	p50	avg	p99	min	samples
size tick: unlabelled (2000 slots)	tick	570 µs	598 µs	1.08 ms	547 µs	1150
size tick: labelled, center/center (re-anchor early-out) (2000 slots)	tick	605 µs	637 µs	1.01 ms	579 µs	1099
size tick: labelled, hanging (sidecar rewrite per tick) (2000 slots)	tick	1.25 ms	1.34 ms	2.27 ms	1.10 ms	505
edge width tick (4000 slots)	bare	277 µs	287 µs	413 µs	267 µs	2427
edge width tick (4000 slots)	with rides (casing + match-line arrows)	690 µs	719 µs	1.10 ms	676 µs	971
padding tick + auto-bounds flush (250 parents × 8 children)	tick + flush	176 µs	186 µs	380 µs	172 µs	3760
font-size tick (2000 labels)	wrap none (dims scale-patch)	368 µs	407 µs	928 µs	345 µs	1699
font-size tick (2000 labels)	wrapped (re-break per tick)	5.69 ms	5.78 ms	6.48 ms	5.38 ms	116
baseline: paint (opacity) tick (2000 slots)	tick	141 µs	145 µs	244 µs	138 µs	4799