Scientists at the University of Glasgow have focused on improving the search speed, hoping to reduce energy consumption by increasing the speed. Dr. will Vanderbilt of the research team pointed out that people will consider the carbon cost of running information technology. They may be very aware of the power consumption of using computers, but they may not pay attention to the energy consumption when using online search. Although there are many ways to reduce the energy consumption of data centers, reducing search time is their focus. By increasing the search speed, the energy consumed by the server under the same workload will be reduced.
The research team used Xilinx FPGAs to perform information retrieval and database document filtering algorithms. The results show that the search speed of FPGA system is 20 times faster than that of Intel An Teng -2 dual-core processor. Moreover, the power consumption of FPGA chips is very small, and each chip only needs to consume 1.25 watts, while the Itanium processor needs to consume 130 watts.
Dr. Azwopadi, another member of the research team, said that FPGAs can reduce energy consumption and improve computing speed, which has great potential and is expected to bring huge economic benefits. If several FPGAs are connected to a processor, the operation speed will be greatly improved, leaving only a few carbon footprints.
Scientists hope to further develop the system and finally apply it to the network server, so as to speed up the network search and reduce the energy and carbon costs required by the search engine. At present, they are working hard to further improve the performance of the prototype and test the data center environment.
Server virtualization
Speech on poverty and wealth 1
Green is a precious wealth endowed by nature, and green is the cradle of human civilization. Everyone is eager