github/tfhe-rs
1
1
Fork 0
mirror of https://github.com/zama-ai/tfhe-rs.git synced 2026-01-09 14:47:56 -05:00
Code Issues Packages Projects Releases Wiki Activity

chore(docs): make difference between benchmarks stand out more

Browse Source
This commit is contained in:
Arthur Meyre
2025-10-23 10:39:35 +02:00
committed by IceTDrinker
parent 95593b1ea9
commit 9d31e994aa
3 changed files with 12 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tfhe/docs/getting-started/benchmarks/cpu/cpu-integer-operations.md
View File

@@ -14,21 +14,21 @@ The following tables benchmark the execution time of some operation sets using `
The next tables show the operation timings on CPU when all inputs are encrypted:
### Latency
### Encrypted/encrypted operations latency
![](../../../.gitbook/assets/cpu-integer-benchmark-tuniform-2m128-ciphertext.svg)
### Throughput
### Encrypted/encrypted operations throughput
![](../../../.gitbook/assets/cpu-integer-benchmark-tuniform-2m128-ciphertext-throughput.svg)
The next tables show the operation timings on CPU when the left input is encrypted and the right is a clear scalar of the same size:
### Latency
### Encrypted/clear operations latency
![](../../../.gitbook/assets/cpu-integer-benchmark-tuniform-2m128-plaintext.svg)
### Throughput
### Encrypted/clear operations throughput
![](../../../.gitbook/assets/cpu-integer-benchmark-tuniform-2m128-plaintext-throughput.svg)

8
tfhe/docs/getting-started/benchmarks/gpu/gpu-integer-operations.md
View File

@@ -13,21 +13,21 @@ The cryptographic parameters `PARAM_GPU_MULTI_BIT_GROUP_4_MESSAGE_2_CARRY_2_KS_P
Below come the results for the execution on eight H100 using SXM technology.
The following tables show the performance when the inputs of the benchmarked operation are encrypted:
### Latency
### Encrypted/encrypted operations latency
![](../../../.gitbook/assets/gpu-integer-benchmark-h100x8-sxm5-multi-bit-tuniform-2m128-ciphertext.svg)
### Throughput
### Encrypted/encrypted operations throughput
![](../../../.gitbook/assets/gpu-integer-benchmark-h100x8-sxm5-multi-bit-tuniform-2m128-ciphertext-throughput.svg)
The following tables show the performance when the left input of the benchmarked operation is encrypted and the other is a clear scalar of the same size:
### Latency
### Encrypted/clear operations latency
![](../../../.gitbook/assets/gpu-integer-benchmark-h100x8-sxm5-multi-bit-tuniform-2m128-plaintext.svg)
### Throughput
### Encrypted/clear operations throughput
![](../../../.gitbook/assets/gpu-integer-benchmark-h100x8-sxm5-multi-bit-tuniform-2m128-plaintext-throughput.svg)

8
tfhe/docs/getting-started/benchmarks/hpu/hpu-integer-operations.md
View File

@@ -13,21 +13,21 @@ Below are the results for the execution on a single Alveo v80 board.
The following tables show the performance when the inputs of the benchmarked operation are encrypted:
### Latency
### Encrypted/encrypted operations latency
![](../../../.gitbook/assets/hpu-integer-benchmark-hpux1-tuniform-2m128-ciphertext.svg)
### Throughput
### Encrypted/encrypted operations throughput
![](../../../.gitbook/assets/hpu-integer-benchmark-hpux1-tuniform-2m128-ciphertext-throughput.svg)
The following tables show the performance when the left input of the benchmarked operation is encrypted and the other is a clear scalar of the same size:
### Latency
### Encrypted/clear operations latency
![](../../../.gitbook/assets/hpu-integer-benchmark-hpux1-tuniform-2m128-plaintext.svg)
### Throughput
### Encrypted/clear operations throughput
![](../../../.gitbook/assets/hpu-integer-benchmark-hpux1-tuniform-2m128-plaintext-throughput.svg)