Data Processing and Information Classification-An In-Memory Approach
| Autor: | Giulia Santoro, Mariagrazia Graziano, Fabrizio Ottati, Marco Vacca, Andrea Marchesin, Milena Andrighetti, Massimo Ruo Roch, Giovanna Turvani, Maurizio Zamboni |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
processing in memory
Computer science Big data Real-time computing 02 engineering and technology lcsh:Chemical technology bitmap indexing 01 natural sciences Biochemistry Article Analytical Chemistry big data 0103 physical sciences 0202 electrical engineering electronic engineering information engineering lcsh:TP1-1185 Electronics Electrical and Electronic Engineering Instrumentation 010302 applied physics Bitmap indexing Internet of things Memory wall Processing in memory Data processing business.industry Process (computing) internet of things Atomic and Molecular Physics and Optics 020202 computer hardware & architecture Classified information memory wall CMOS Server farm electrical_electronic_engineering Hardware acceleration business |
| Zdroj: | Sensors (Basel, Switzerland) Sensors Volume 20 Issue 6 Sensors, Vol 20, Iss 6, p 1681 (2020) |
| ISSN: | 1424-8220 |
| Popis: | To live in the information society means to be surrounded by billions of electronic devices full of sensors that constantly acquire data. This enormous amount of data must be processed and classified. A solution commonly adopted is to send these data to server farms to be remotely elaborated. The drawback is a huge battery drain due to high amount of information that must be exchanged. To compensate this problem data must be processed locally, near the sensor itself. But this solution requires huge computational capabilities. While microprocessors, even mobile ones, nowadays have enough computational power, their performance are severely limited by the Memory Wall problem. Memories are too slow, so microprocessors cannot fetch enough data from them, greatly limiting their performance. A solution is the Processing-In-Memory (PIM) approach. New memories are designed that are able to elaborate data inside them eliminating the Memory Wall problem. In this work we present an example of such system, using as a case of study the Bitmap Indexing algorithm. Such algorithm is used to classify data coming from many sources in parallel. We propose an hardware accelerator designed around the Processing-In-Memory approach, that is capable of implementing this algorithm and that can also be reconfigured to do other tasks or to work as standard memory. The architecture has been synthesized using CMOS technology. The results that we have obtained highlights that, not only it is possible to process and classify huge amount of data locally, but also that it is possible to obtain this result with a very low power consumption. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|