Benchmarking ProtoGalaxy #692
Assignees
Milestone
Comments
codygunton
added a commit
to AztecProtocol/aztec-packages
that referenced
this issue
Jan 11, 2024
Benchmark Protogalaxy prover. This just benchmarks the current folding function. Future PRs will make update and augment with values that will better reflect performance in production. Results: uses 16 cores (for the x86 build). `fold_one/N` measures the time to call the `fold_instances` function to fold a single instance into an accumulator where the instance sizes are $2^N$. Note this is for the Ultra arithmetization (so no Goblin, Poseidon2 or DataBus relations). ``` % taskset -c 0-15 ./bin/protogalaxy_bench 38s ~/barretenberg-cpp/build cg/pg-bench mainframe 2024-01-11T19:39:49+00:00 Running ./bin/protogalaxy_bench Run on (128 X 2649.99 MHz CPU s) CPU Caches: L1 Data 32 KiB (x64) L1 Instruction 32 KiB (x64) L2 Unified 512 KiB (x64) L3 Unified 32768 KiB (x8) Load Average: 3.48, 8.51, 27.24 ------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------ fold_one/14 417 ms 259 ms 3 fold_one/15 668 ms 409 ms 2 fold_one/16 1158 ms 696 ms 1 fold_one/17 2245 ms 1433 ms 1 fold_one/18 4168 ms 2644 ms 1 fold_one/19 7890 ms 5119 ms 1 fold_one/20 15809 ms 10633 ms 1 ``` Closes AztecProtocol/barretenberg#692
AztecBot
pushed a commit
that referenced
this issue
Jan 12, 2024
Benchmark Protogalaxy prover. This just benchmarks the current folding function. Future PRs will make update and augment with values that will better reflect performance in production. Results: uses 16 cores (for the x86 build). `fold_one/N` measures the time to call the `fold_instances` function to fold a single instance into an accumulator where the instance sizes are $2^N$. Note this is for the Ultra arithmetization (so no Goblin, Poseidon2 or DataBus relations). ``` % taskset -c 0-15 ./bin/protogalaxy_bench 38s ~/barretenberg-cpp/build cg/pg-bench mainframe 2024-01-11T19:39:49+00:00 Running ./bin/protogalaxy_bench Run on (128 X 2649.99 MHz CPU s) CPU Caches: L1 Data 32 KiB (x64) L1 Instruction 32 KiB (x64) L2 Unified 512 KiB (x64) L3 Unified 32768 KiB (x8) Load Average: 3.48, 8.51, 27.24 ------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------ fold_one/14 417 ms 259 ms 3 fold_one/15 668 ms 409 ms 2 fold_one/16 1158 ms 696 ms 1 fold_one/17 2245 ms 1433 ms 1 fold_one/18 4168 ms 2644 ms 1 fold_one/19 7890 ms 5119 ms 1 fold_one/20 15809 ms 10633 ms 1 ``` Closes #692
michaelelliot
pushed a commit
to Swoir/noir_rs
that referenced
this issue
Feb 28, 2024
Benchmark Protogalaxy prover. This just benchmarks the current folding function. Future PRs will make update and augment with values that will better reflect performance in production. Results: uses 16 cores (for the x86 build). `fold_one/N` measures the time to call the `fold_instances` function to fold a single instance into an accumulator where the instance sizes are $2^N$. Note this is for the Ultra arithmetization (so no Goblin, Poseidon2 or DataBus relations). ``` % taskset -c 0-15 ./bin/protogalaxy_bench 38s ~/barretenberg-cpp/build cg/pg-bench mainframe 2024-01-11T19:39:49+00:00 Running ./bin/protogalaxy_bench Run on (128 X 2649.99 MHz CPU s) CPU Caches: L1 Data 32 KiB (x64) L1 Instruction 32 KiB (x64) L2 Unified 512 KiB (x64) L3 Unified 32768 KiB (x8) Load Average: 3.48, 8.51, 27.24 ------------------------------------------------------ Benchmark Time CPU Iterations ------------------------------------------------------ fold_one/14 417 ms 259 ms 3 fold_one/15 668 ms 409 ms 2 fold_one/16 1158 ms 696 ms 1 fold_one/17 2245 ms 1433 ms 1 fold_one/18 4168 ms 2644 ms 1 fold_one/19 7890 ms 5119 ms 1 fold_one/20 15809 ms 10633 ms 1 ``` Closes AztecProtocol/barretenberg#692
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
No description provided.
The text was updated successfully, but these errors were encountered: