Tremendous progress has been made in building computer systems with higher performance over the last several decades. However the need to build systems with even higher performance exists in a number of key strategic areas. CMOS scaling will continue to be exploited to increase the areal density of devices and to reduce the power per operation beyond the 10nm node. For some applications like training neuromorphic systems and the modeling of quantum systems even more performance is required. Some of the increased performance can be obtained through re-architecting the system, using lower precision computations, via GPU like architectures with more inter thread communication bandwidth, and using FPGA like programmability. Beyond this device level innovation is required. We will review some of the progress that has been made in developing the materials and processes to enable performance increases at the device level.
1:00 p.m. – 2:00 p.m. (Gaithersburg, Bldg. 101, Lecture Rm. B)
11:00 a.m. – 12:00 p.m. (Boulder, VTC in 81 1116)
Robert Wisnieff, Distinguished Research Staff Member
IBM TJ Watson Research Center