The large energy cost of memory fetches limits the overallefficiency of applications no matter how efficient the ac-celerators are on the chip. As a result the most importantoptimization must be done at the algorithm level, to reduce off-chip memory accesses, to createDark Memory. The algorithmsmust first be (re)written for both locality and parallelism beforeyou tailor the hardware to accelerate them.Using Pareto curves in theenergy/opandmm2/(op/s)spaceallows one to quickly evaluate different accelerators, memorysystems, and even algorithms to understand the trade-offsbetween performance, power and die area. This analysis isa powerful way to optimize chips in the Dark Silicon era.
CS311 GDB 1 Solution and Discussion
CS311 GDB – 2020-2021
Monday, February 15, 2021
Monday, February 22, 2021
Due to the current COVID-19 Pandemic NCOC (National Command and Operation Center) decides to develop a single national level digital platform where all the information regarding COVID Pandemic will be available nationwide. The aim of this digital platform is to provide authentic and verified information at national level.
Suppose you are working in a firm and NCOC takes the services of that firm for the creation of this digital platform. You are assigned the task to develop a web service for this web application. NCOC wants that the web service should be platform independent and should be able to entertain all the HTTP requests. You are given the choice to develop this web service using any one of the following technologies.
· Java Servlets
· CGI (Common Gateway Interface)
Which technology you will consider for the creatin of this web service for NCOC digital platform and why? Support your answers with some solid reason.