Skip to content

Improve CSP#1838

Open
HayimShaul wants to merge 26 commits into
mainfrom
hayim_temp_jun25
Open

Improve CSP#1838
HayimShaul wants to merge 26 commits into
mainfrom
hayim_temp_jun25

Conversation

@HayimShaul

@HayimShaul HayimShaul commented Jul 1, 2026

Copy link
Copy Markdown
Contributor

Improve the CSP ZKP

co-authored with: @kushnireyal

@HayimShaul HayimShaul added this to the Q3/26 milestone Jul 1, 2026
@HayimShaul HayimShaul self-assigned this Jul 1, 2026
@HayimShaul
HayimShaul requested a review from adecaro July 1, 2026 12:50
@adecaro

adecaro commented Jul 1, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Hi @HayimShaul , thanks for this effort. It looks promising. Please, fix the errors reported by the CI so we can check if the tests are passing. Please, also run the benchmarks to understand which improvements this PR brings in.
Thanks 🙏

@neetance , you might be interested in having a look at this as well. Thanks 🙏

@HayimShaul
HayimShaul marked this pull request as ready for review July 1, 2026 15:00
@adecaro

adecaro commented Jul 2, 2026

Copy link
Copy Markdown
Contributor

Hi @HayimShaul , please, run the benchmark with and without the optimization to see what's the gain. Please, post the result here. Thanks 🙏

@AkramBitar

AkramBitar commented Jul 2, 2026

Copy link
Copy Markdown
Contributor

@HayimShaul could you please open an issue and attached it to this PR. Please put the issue under the generic EPIC.

HayimShaul and others added 11 commits July 2, 2026 12:12
@HayimShaul
optimize lagrange numerator
72b5b61 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
reduce alloc size
f636de6 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
dont copy leaves at 2nd step
853084d 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
no special treetment for root in 2nd step
6efaa35 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@kushnireyal @HayimShaul
optimize CSP verify: fold padding generators and merge final MSMs (#1833
60c9a59 
)

Signed-off-by: Eyal Kushnir <Eyal.Kushnir@ibm.com>
Co-authored-by: Eyal Kushnir <Eyal.Kushnir@ibm.com>
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
fmt
710320a 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
lint
7305100 
Signed-off-by: Hayim.Shaul@ibm.com <hayimsha@fhe03.vpc.cloud9.ibm.com>
@HayimShaul
fmt
cd5c02b 
Signed-off-by: Hayim.Shaul@ibm.com <hayimsha@fhe03.vpc.cloud9.ibm.com>
@HayimShaul
added tests
1e47f45 
Signed-off-by: Hayim.Shaul@ibm.com <hayimsha@fhe03.vpc.cloud9.ibm.com>
@HayimShaul
lint
d26f6c1 
Signed-off-by: Hayim.Shaul@ibm.com <hayimsha@fhe03.vpc.cloud9.ibm.com>
@HayimShaul
fmt
f1f90f0 
Signed-off-by: Hayim.Shaul@ibm.com <hayimsha@fhe03.vpc.cloud9.ibm.com>
@HayimShaul
turn off GC in benchmarks
0215a2c 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
pool trees
45e7420 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
cleanup
7e48a58 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
bug fix
4a30253 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
cleanup
f312f1a 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
cleanup
d54cc7b 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
cleanup
c30700c 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
Revert "turn off GC in benchmarks"
340d5ff 
This reverts commit 0215a2c.

Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
HayimShaul and others added 2 commits July 13, 2026 06:31
@HayimShaul
fmt
d7ca520 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
lint
b818c36 
Signed-off-by: Hayim.Shaul@ibm.com <hayimsha@fhe03.vpc.cloud9.ibm.com>
@HayimShaul HayimShaul linked an issue Jul 13, 2026 that may be closed by this pull request
Improve lagrange numerators generation in zero knowledge proofs #1899
Open
@HayimShaul
clean
1334b7e 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
better code
013c3ba 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
added largrange tests
2638295 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul
add pairs optimization
6cece80 
Signed-off-by: Hayim Shaul <hayim.shaul@gmail.com>
@HayimShaul

HayimShaul commented Jul 15, 2026

Copy link
Copy Markdown
Contributor Author

Here are the benchmarks

IPA
Real Throughput: 846.80/s -> 908.85/s (+7.3% improvement)
Pure Throughput: 847.04/s -> 909.07/s (+7.3% improvement)
Allocs: 9079 Allocs/op -> 9126 Allocs/op (+0.5% worse - practically the same)
Average: 75.55ms -> 70.40ms (6.9% faster)
P50: 66.20ms -> 65.02ms (1.8% faster)

CSP
Real Throughput: 380.69/s -> 450.57/s (+18.3% improvement)
Pure Throughput: 380.72/s -> 450.62/s (+18.3% improvement)
Allocs: 6968 Allocs/op -> 6496 Allocs/op (+6.77% improvement)
Average: 26.26ms -> 22.19ms (+15.5% improvement)
P50: 25.88ms -> 21.83ms (+15.6% improvement)

The raw benchmark output:

IPA before optimizations

go test -v ./token/core/zkatdlog/nogh/v1/crypto/rp/bulletproof/... -run TestParallelIPAProver -count=1 -args -duration=300s
=== RUN   TestParallelIPAProver
=== RUN   TestParallelIPAProver/Setup(bits_32,_curve_BN254,_#i_2,_#o_2)_with_64_workers
Metric           Value     Description
------           -----     -----------
Workers          64        
Total Ops        254097    (Robust Sample)
Duration         5m0.066s  (Good Duration)
Real Throughput  846.80/s  Observed Ops/sec (Wall Clock)
Pure Throughput  847.04/s  Theoretical Max (Low Overhead)

Latency Distribution:
 Min           17.114115ms   
 P50 (Median)  66.206006ms   
 Average       75.557523ms   
 P95           136.106531ms  
 P99           258.92779ms   
 P99.9         606.836611ms  
 Max           1.467003162s  (Stable Tail)

Stability Metrics:
 Std Dev  48.678587ms  
 IQR      40.385421ms  Interquartile Range
 Jitter   34.772684ms  Avg delta per worker
 CV       64.43%       Unstable (>20%) - Result is Noisy

System Health & Reliability:
 Error Rate          0.0000%          (100% Success) (0 errors)
 Memory              987112 B/op      Allocated bytes per operation
 Allocs              9079 allocs/op   Allocations per operation
 Alloc Rate          724.88 MB/s      Memory pressure on system
 GC Overhead         23.78%           (Severe GC Thrashing)
 GC Pause            1m11.361744841s  Total Stop-The-World time
 GC Cycles           19570            Full garbage collection cycles
 Goroutines Created  1496             Net goroutines above baseline during recording

Latency Heatmap (Dynamic Range):
Range                       Freq   Distribution Graph
 17.114115ms-21.380028ms    76      (0.0%)
 21.380028ms-26.709275ms    943     (0.4%)
 26.709275ms-33.366905ms    9537   ███████ (3.8%)
 33.366905ms-41.684036ms    30925  ████████████████████████ (12.2%)
 41.684036ms-52.074318ms    31483  █████████████████████████ (12.4%)
 52.074318ms-65.054512ms    49953  ████████████████████████████████████████ (19.7%)
 65.054512ms-81.270186ms    47876  ██████████████████████████████████████ (18.8%)
 81.270186ms-101.527826ms   41205  ████████████████████████████████ (16.2%)
 101.527826ms-126.834942ms  24871  ███████████████████ (9.8%)
 126.834942ms-158.450184ms  10326  ████████ (4.1%)
 158.450184ms-197.94593ms   3015   ██ (1.2%)
 197.94593ms-247.2865ms     1147    (0.5%)
 247.2865ms-308.925841ms    871     (0.3%)
 308.925841ms-385.929581ms  738     (0.3%)
 385.929581ms-482.127494ms  533     (0.2%)
 482.127494ms-602.303974ms  331     (0.1%)
 602.303974ms-752.435988ms  169     (0.1%)
 752.435988ms-939.990337ms  72      (0.0%)
 939.990337ms-1.174295021s  20      (0.0%)
 1.174295021s-1.467003162s  6       (0.0%)

--- Analysis & Recommendations ---
[FAIL] High Variance (CV 64.43%). System noise is affecting results. Isolate the machine or increase duration.
[INFO] High Allocations (9079/op). This will trigger frequent GC cycles and increase Max Latency.
----------------------------------

--- Throughput Timeline ---
Timeline: [▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇█▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▆▆▇▆▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▃▂ ▂▂▂▅▇▆▃▂▂ ▂▃▂▂▂▂▂▂▂▃▃▂▂▂▅▅▆▆▆▅▆▅] (Max: 978 ops/s)

--- PASS: TestParallelIPAProver (303.28s)
    --- PASS: TestParallelIPAProver/Setup(bits_32,_curve_BN254,_#i_2,_#o_2)_with_64_workers (303.28s)
PASS
ok      github.com/LFDT-Panurus/panurus/token/core/zkatdlog/nogh/v1/crypto/rp/bulletproof       303.334s

IPA after optimizations

go test -v ./token/core/zkatdlog/nogh/v1/crypto/rp/bulletproof/... -run TestParallelIPAProver -count=1 -args -duration=300s
=== RUN   TestParallelIPAProver
=== RUN   TestParallelIPAProver/Setup(bits_32,_curve_BN254,_#i_2,_#o_2)_with_64_workers
Metric           Value     Description
------           -----     -----------
Workers          64        
Total Ops        272698    (Robust Sample)
Duration         5m0.048s  (Good Duration)
Real Throughput  908.85/s  Observed Ops/sec (Wall Clock)
Pure Throughput  909.07/s  Theoretical Max (Low Overhead)

Latency Distribution:
 Min           16.274249ms   
 P50 (Median)  65.023249ms   
 Average       70.401678ms   
 P95           125.222898ms  
 P99           156.730353ms  
 P99.9         196.880151ms  
 Max           352.952508ms  (Stable Tail)

Stability Metrics:
 Std Dev  28.978657ms  
 IQR      38.624345ms  Interquartile Range
 Jitter   31.696379ms  Avg delta per worker
 CV       41.16%       Unstable (>20%) - Result is Noisy

System Health & Reliability:
 Error Rate          0.0000%          (100% Success) (0 errors)
 Memory              1006915 B/op     Allocated bytes per operation
 Allocs              9126 allocs/op   Allocations per operation
 Alloc Rate          778.07 MB/s      Memory pressure on system
 GC Overhead         25.63%           (Severe GC Thrashing)
 GC Pause            1m16.890971022s  Total Stop-The-World time
 GC Cycles           22439            Full garbage collection cycles
 Goroutines Created  1433             Net goroutines above baseline during recording

Latency Heatmap (Dynamic Range):
Range                       Freq   Distribution Graph
 16.274249ms-18.980678ms    11      (0.0%)
 18.980678ms-22.137189ms    112     (0.0%)
 22.137189ms-25.818633ms    559     (0.2%)
 25.818633ms-30.112306ms    3391   ███ (1.2%)
 30.112306ms-35.120022ms    14540  ██████████████ (5.3%)
 35.120022ms-40.960528ms    25639  ████████████████████████ (9.4%)
 40.960528ms-47.772318ms    22240  █████████████████████ (8.2%)
 47.772318ms-55.716917ms    28455  ███████████████████████████ (10.4%)
 55.716917ms-64.982714ms    41231  ████████████████████████████████████████ (15.1%)
 64.982714ms-75.789425ms    36229  ███████████████████████████████████ (13.3%)
 75.789425ms-88.393306ms    35757  ██████████████████████████████████ (13.1%)
 88.393306ms-103.093229ms   28129  ███████████████████████████ (10.3%)
 103.093229ms-120.237769ms  19196  ██████████████████ (7.0%)
 120.237769ms-140.233469ms  10691  ██████████ (3.9%)
 140.233469ms-163.554479ms  4684   ████ (1.7%)
 163.554479ms-190.753804ms  1447   █ (0.5%)
 190.753804ms-222.476413ms  330     (0.1%)
 222.476413ms-259.474534ms  51      (0.0%)
 259.474534ms-302.62549ms   4       (0.0%)
 302.62549ms-352.952508ms   2       (0.0%)

--- Analysis & Recommendations ---
[FAIL] High Variance (CV 41.16%). System noise is affecting results. Isolate the machine or increase duration.
[INFO] High Allocations (9126/op). This will trigger frequent GC cycles and increase Max Latency.
----------------------------------

--- Throughput Timeline ---
Timeline: [▆▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇█▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇] (Max: 986 ops/s)

--- PASS: TestParallelIPAProver (303.25s)
    --- PASS: TestParallelIPAProver/Setup(bits_32,_curve_BN254,_#i_2,_#o_2)_with_64_workers (303.25s)
PASS
ok      github.com/LFDT-Panurus/panurus/token/core/zkatdlog/nogh/v1/crypto/rp/bulletproof       303.296s

CSP before optimizations

go test ./token/core/zkatdlog/nogh/v1/validator -test.run=TestParallelBenchmarkValidatorTransfer -test.v -test.timeout 0 -bits=32 -curves=BLS12_381_BBS_GURVY -num_inputs=2 -num_outputs=2 -workers=10 -duration=300s -setup_samples=128 -executor=serial -proof_type=csp
=== RUN   TestParallelBenchmarkValidatorTransfer
=== RUN   TestParallelBenchmarkValidatorTransfer/Setup(bits_32,_curve_BLS12_381_BBS_GURVY,_#i_2,_#o_2)_with_10_workers
Metric           Value     Description
------           -----     -----------
Workers          10        
Total Ops        114215    (Robust Sample)
Duration         5m0.021s  (Good Duration)
Real Throughput  380.69/s  Observed Ops/sec (Wall Clock)
Pure Throughput  380.72/s  Theoretical Max (Low Overhead)

Latency Distribution:
 Min           19.091819ms  
 P50 (Median)  25.889035ms  
 Average       26.265791ms  
 P95           30.981896ms  
 P99           33.105655ms  
 P99.9         35.145516ms  
 Max           40.170126ms  (Stable Tail)

Stability Metrics:
 Std Dev  2.5196ms    
 IQR      3.406522ms  Interquartile Range
 Jitter   2.779699ms  Avg delta per worker
 CV       9.59%       (Acceptable 5-10%)

System Health & Reliability:
 Error Rate          0.0000%         (100% Success) (0 errors)
 Memory              743014 B/op     Allocated bytes per operation
 Allocs              6968 allocs/op  Allocations per operation
 Alloc Rate          260.11 MB/s     Memory pressure on system
 GC Overhead         7.59%           (Severe GC Thrashing)
 GC Pause            22.76209739s    Total Stop-The-World time
 GC Cycles           25818           Full garbage collection cycles
 Goroutines Created  46              Net goroutines above baseline during recording

Latency Heatmap (Dynamic Range):
Range                     Freq   Distribution Graph
 19.091819ms-19.815274ms  41      (0.0%)
 19.815274ms-20.566145ms  194     (0.2%)
 20.566145ms-21.345468ms  555    █ (0.5%)
 21.345468ms-22.154323ms  1470   ███ (1.3%)
 22.154323ms-22.993828ms  4748   ██████████ (4.2%)
 22.993828ms-23.865144ms  11652  █████████████████████████ (10.2%)
 23.865144ms-24.769478ms  16996  █████████████████████████████████████ (14.9%)
 24.769478ms-25.708081ms  18121  ████████████████████████████████████████ (15.9%)
 25.708081ms-26.68225ms   16682  ████████████████████████████████████ (14.6%)
 26.68225ms-27.693334ms   13819  ██████████████████████████████ (12.1%)
 27.693334ms-28.742731ms  10867  ███████████████████████ (9.5%)
 28.742731ms-29.831894ms  7990   █████████████████ (7.0%)
 29.831894ms-30.962329ms  5293   ███████████ (4.6%)
 30.962329ms-32.135601ms  3159   ██████ (2.8%)
 32.135601ms-33.353331ms  1702   ███ (1.5%)
 33.353331ms-34.617206ms  685    █ (0.6%)
 34.617206ms-35.928973ms  199     (0.2%)
 35.928973ms-37.290448ms  35      (0.0%)
 37.290448ms-38.703514ms  5       (0.0%)
 38.703514ms-40.170126ms  2       (0.0%)

--- Analysis & Recommendations ---
[INFO] High Allocations (6968/op). This will trigger frequent GC cycles and increase Max Latency.
[PASS] RunBenchmark looks healthy and statistically sound.
----------------------------------

--- Throughput Timeline ---
Timeline: [▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇█▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇] (Max: 398 ops/s)

--- PASS: TestParallelBenchmarkValidatorTransfer (343.80s)
    --- PASS: TestParallelBenchmarkValidatorTransfer/Setup(bits_32,_curve_BLS12_381_BBS_GURVY,_#i_2,_#o_2)_with_10_workers (343.77s)
PASS
ok      github.com/LFDT-Panurus/panurus/token/core/zkatdlog/nogh/v1/validator   343.850s

CSP after optimizations

go test ./token/core/zkatdlog/nogh/v1/validator -test.run=TestParallelBenchmarkValidatorTransfer -test.v -test.timeout 0 -bits=32 -curves=BLS12_381_BBS_GURVY -num_inputs=2 -num_outputs=2 -workers=10 -duration=300s -setup_samples=128 -executor=serial -proof_type=csp
=== RUN   TestParallelBenchmarkValidatorTransfer
=== RUN   TestParallelBenchmarkValidatorTransfer/Setup(bits_32,_curve_BLS12_381_BBS_GURVY,_#i_2,_#o_2)_with_10_workers
Metric           Value     Description
------           -----     -----------
Workers          10        
Total Ops        135178    (Robust Sample)
Duration         5m0.015s  (Good Duration)
Real Throughput  450.57/s  Observed Ops/sec (Wall Clock)
Pure Throughput  450.62/s  Theoretical Max (Low Overhead)

Latency Distribution:
 Min           16.498301ms  
 P50 (Median)  21.830298ms  
 Average       22.191834ms  
 P95           26.847587ms  
 P99           29.160549ms  
 P99.9         31.498109ms  
 Max           34.770375ms  (Stable Tail)

Stability Metrics:
 Std Dev  2.475655ms  
 IQR      3.12062ms   Interquartile Range
 Jitter   2.683202ms  Avg delta per worker
 CV       11.16%      Moderate Variance (10-20%)

System Health & Reliability:
 Error Rate          0.0000%         (100% Success) (0 errors)
 Memory              639948 B/op     Allocated bytes per operation
 Allocs              6496 allocs/op  Allocations per operation
 Alloc Rate          255.96 MB/s     Memory pressure on system
 GC Overhead         6.30%           (Severe GC Thrashing)
 GC Pause            18.915443876s   Total Stop-The-World time
 GC Cycles           23672           Full garbage collection cycles
 Goroutines Created  82              Net goroutines above baseline during recording

Latency Heatmap (Dynamic Range):
Range                     Freq   Distribution Graph
 16.498301ms-17.124887ms  22      (0.0%)
 17.124887ms-17.775271ms  920    █ (0.7%)
 17.775271ms-18.450355ms  4518   ████████ (3.3%)
 18.450355ms-19.151079ms  7694   ███████████████ (5.7%)
 19.151079ms-19.878415ms  9353   ██████████████████ (6.9%)
 19.878415ms-20.633374ms  13961  ███████████████████████████ (10.3%)
 20.633374ms-21.417006ms  20178  ███████████████████████████████████████ (14.9%)
 21.417006ms-22.2304ms    20512  ████████████████████████████████████████ (15.2%)
 22.2304ms-23.074685ms    15971  ███████████████████████████████ (11.8%)
 23.074685ms-23.951035ms  12540  ████████████████████████ (9.3%)
 23.951035ms-24.860668ms  9917   ███████████████████ (7.3%)
 24.860668ms-25.804848ms  7412   ██████████████ (5.5%)
 25.804848ms-26.784887ms  5173   ██████████ (3.8%)
 26.784887ms-27.802146ms  3320   ██████ (2.5%)
 27.802146ms-28.85804ms   1976   ███ (1.5%)
 28.85804ms-29.954035ms   1017   █ (0.8%)
 29.954035ms-31.091655ms  479     (0.4%)
 31.091655ms-32.272481ms  160     (0.1%)
 32.272481ms-33.498153ms  44      (0.0%)
 33.498153ms-34.770375ms  11      (0.0%)

--- Analysis & Recommendations ---
[INFO] High Allocations (6496/op). This will trigger frequent GC cycles and increase Max Latency.
----------------------------------

--- Throughput Timeline ---
Timeline: [▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇█▇▇█▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇] (Max: 474 ops/s)

--- PASS: TestParallelBenchmarkValidatorTransfer (344.41s)
    --- PASS: TestParallelBenchmarkValidatorTransfer/Setup(bits_32,_curve_BLS12_381_BBS_GURVY,_#i_2,_#o_2)_with_10_workers (344.38s)
PASS
ok      github.com/LFDT-Panurus/panurus/token/core/zkatdlog/nogh/v1/validator   344.466s

fmt
95b4c4c 
Signed-off-by: Hayim.Shaul@ibm.com <hayimsha@fhe03.vpc.cloud9.ibm.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@adecaro adecaro Awaiting requested review from adecaro

At least 1 approving review is required to merge this pull request.

Assignees

@HayimShaul HayimShaul

Labels

benchmark

Projects

None yet

Milestone

Q3/26

Development

Successfully merging this pull request may close these issues.

Improve lagrange numerators generation in zero knowledge proofs

4 participants

@HayimShaul @adecaro @AkramBitar @kushnireyal