A modern computing system is an elaborate stack of hardware and software with stable intermediate points that enable a platform builder to construct all sorts of machines with ‘commodity’ hardware and software pieces. Unfortunately, this extraordinary flexibility comes at a huge cost: two to three orders of magnitude in efficiency and cost. This is far beyond the range of most engineered systems. This observation has been at the root of our quest for new architectures — from modifications to von Neumann to data-flow variants — and new ways of building/synthesizing hardware pieces for new machines — co-processors to synthesized accelerators. This talk pulls the most interesting and promising vectors of research in this line of thinking to outline a vision of emerging machines: that are engineered less but enabled more to sense and adapt the computation to the environment in which they are placed.
Rajesh K. Gupta is a QUALCOMM professor in Computer Science and Engineering at UC San Diego. His research focus is on timing and energy issues in embedded systems. His past contributions include SystemC modeling and SPARK parallelizing high-level synthesis, both of which have been incorporated into industrial practice. Earlier Gupta lead or co-lead DARPA-sponsored efforts under the Data Intensive Systems (DIS) and Power Aware Computing and Communications (PACC) programs on role of adaptation in energy efficient system architectures. His ongoing projects are focused on mitigating microelectronic variability and improving precision of time in embedded systems. In recent years, Gupta and his students have received a best paper award at IEEE/ACM DCOSS’08 and a best demonstration award at IEEE/ACM IPSN/SPOTS’05. Gupta received a BTech in EE from IIT Kanpur, MS in EECS from UC Berkeley and a PhD in Electrical Engineering from Stanford University. Gupta is a Fellow of the IEEE. Gupta serves as chair of Computer Science and as associate director of the Qualcomm Institute at UCSD.