Intel has announced a partnership with fabless chip firm eASIC to develop Xeon processors with customisable logic that will enable customers to optimise the chips for specific workloads such as security and big data.
The collaboration is part of Intel's strategy to integrate programmable technology with Xeon processors to give service provider customers and others custom solutions that lower cost and greatly improve performance and power.
Intel said that the new parts will enable acceleration up to twice the speed of a field programmable gate array (FPGA) for workloads like security and big data analytics, while cutting the time to market for custom application specific integrated circuit (ASIC) development by as much as 50 percent.
"Having the ability to highly customise our solutions for a given workload will not only make the specific application run faster, but help accelerate the growth of exciting new applications like visual search," said Diane Bryant, senior vice president of Intel's Data Centre Group.
This is not the first time that Intel has disclosed custom versions of its chips. Bryant detailed plans last year to combine Xeon E5 processors with an FPGA inside the same chip package, so that it could still be used in a standard server motherboard.
Intel did not divulge which chips are getting the custom treatment this time around, saying only that eASIC planned to integrate its technology "with future Intel Xeon processors".
However, Intel said that it had chosen eASIC's technology because it offered increased flexibility and faster time to market compared with traditional ASICs, as well as better performance and lower power consumption compared with standard FPGAs.
In Oracle's case, the chip was an Xeon E7 able dynamically to switch its core count, clock frequency and power consumption without the need for a system-level reboot, while Amazon opted for a chip able to run at higher clock speeds.
However, such customisations are costly, which is why Intel is targeting customers such as cloud service providers, which regularly invest in a lot of new servers and can afford a highly customised, integrated hardware solution if it makes sense for their particular workload.
Could be used for everything from search-and-rescue robots to wearable tech
Don't require the rare material being mined from the mountains of South America
IBM hopes that its new tool will avoid bias in artificial intelligence
Found by calculating the strength of the material deep inside the crust of neutron stars