Neural Net Accelerator

Neural Net Coprocessors

SSRLabs' Neural Net Coprocessors are massively parallel accelerators for Artificial Intelligence and Decision Support systems with an industry-leading performance in learning and unlearning rates. Due to their high-bandwidth interfaces both to memory and to other coprocessors they scale out in a more linear fashion than any other accelerator.

Floating Point Accelerator

Floating-Point Coprocessors

SSRLabs' Floating Point Coprocessors are massively parallel accelerators for traditional HPC such as FEMs and n-body problems with an industry-leading performance. Due to their high-bandwidth interfaces both to memory and to other coprocessors and the fact that they are not constrained to SIMD operation they scale out in a more linear fashion than any other accelerator.

Very Large Capacity Memory

Very Large Capacity Memory

SSRLabs' Very Large Capacity Memory Modules are Hybrid Memory Cube (HMC) based memories with all of the advantages of standard HMC memories with the added benefit of substantially higher densities. Since they can be programmed to be volatile or non-volatile, they can be used in storage appliance applications where in-memory compute is a requirement, with or without a file system.

licensable IP

Licensable IP

While developing SSRLabs' pScale™ Coprocessors we have created non-core IP and synthesizable building blocks that we are interested in licensing out or selling. These building blocks are related to I/O, memory host adapters (Hybrid Memory Cube) and other components.