Benchmarks

To showcase the performance improvements so far, we have done CPU-based benchmarks. MatSum represents an addition of two encrypted matrices, MatMulA represents the multiplication of encrypted with plaintext matrix and MatMulB represents the multiplication of two encrypted matrices. Below we compare CoFHE, Microsoft Seal(CKKS) and Zama's Concrete(tFHE) libraries for input matrix of size 64*64 for MatSum and 8*64 for MatMulA and MatMulB and weight matrix of 64*64(for all). The table shows amortized cost of the operations for just 50 sequential operations. "-" represents memory overflow.

Framework	MatSum Latency	MatMulA Latency	MatMulB Latency
CoFHE	4.06 ms	66.6 ms	4224 ms
Microsoft SEAL	0.04 ms	369.17 s	-
Zama ConcreteML	480ms	52 s	-

Since there are no bootstrapping required the amortised latency of our proposed MatSum and MatMul operations can be significantly reduced by leveraging the parallel computing capabilities of GPUs.