ARM backs embedded MRAM on Samsung’s FDSOI process
ARM has completed its first eMRAM IP test chip tape out and the compiler will be available for use by lead partners in 4Q18, according to Kelvin Low, vice president of marketing for the physical design group at ARM.
Low was formerly senior director of marketing and business development at Samsung Foundry (see FDSOI to get embedded MRAM, flash options at 28nm).
MRAM is a non-volatile memory that can be used to replace flash memory and significantly reduce power consumption compared with systems based on active memory. However, although eFlash has been used for a number of generations of manufacturing it has trouble scaling to the more advanced nodes.
The eMRAM can be integrated with as few as three additional masks, while embedded flash requires greater than 12 additional masks at 40nm and below, Low said.
“Also, the eMRAM compiler can generate instances to replace Flash, Electrically Erasable Programmable Read-Only Memory (EEPROM) and slow SRAM/data buffer memories with a single non-volatile fast memory – particularly suited for cost- and power-sensitive IoT applications,” Low said in an online article.
ARM Artisan physical IP offerings have also been announced for Samsung Foundry’s 11LPP, 7LPP and 5LPE EUV process nodes. ARM 7LPP physical IP platform is available starting 3Q18 for lead partners, while the 5LPE physical IP platform offering will be available in early 2019.
Related links and articles: