Demeter: A Fast and Energy-Efficient Food Profiler Using Hyperdimensional Computing in Memory
Food profiling is an essential step in any food monitoring system needed to prevent health risks and potential frauds in the food industry. Significant improvements in sequencing technologies are pushing food profiling to become the main computational bottleneck. State-of-the-art profilers are unfor...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9847238/ |
| _version_ | 1811216137803792384 |
|---|---|
| author | Taha Shahroodi Mahdi Zahedi Can Firtina Mohammed Alser Stephan Wong Onur Mutlu Said Hamdioui |
| author_facet | Taha Shahroodi Mahdi Zahedi Can Firtina Mohammed Alser Stephan Wong Onur Mutlu Said Hamdioui |
| author_sort | Taha Shahroodi |
| collection | DOAJ |
| description | Food profiling is an essential step in any food monitoring system needed to prevent health risks and potential frauds in the food industry. Significant improvements in sequencing technologies are pushing food profiling to become the main computational bottleneck. State-of-the-art profilers are unfortunately too costly for food profiling. Our goal is to design a food profiler that solves the main limitations of existing profilers, namely (1) working on massive data structures and (2) incurring considerable data movement, for a real-time monitoring system. To this end, we propose Demeter, the first platform-independent framework for food profiling. Demeter overcomes the first limitation through the use of hyperdimensional computing (HDC) and efficiently performs the accurate few-species classification required in food profiling. We overcome the second limitation by the use of an in-memory hardware accelerator for Demeter (named Acc-Demeter) based on memristor devices. Acc-Demeter actualizes several domain-specific optimizations and exploits the inherent characteristics of memristors to improve the overall performance and energy consumption of Acc-Demeter. We compare Demeter’s accuracy with other industrial food profilers using detailed software modeling. We synthesize Acc-Demeter’s required hardware using UMC’s 65nm library by considering an accurate PCM model based on silicon-based prototypes. Our evaluations demonstrate that Acc-Demeter achieves a (1) throughput improvement of <inline-formula> <tex-math notation="LaTeX">$192\times $ </tex-math></inline-formula> and <inline-formula> <tex-math notation="LaTeX">$724\times $ </tex-math></inline-formula> and (2) memory reduction of <inline-formula> <tex-math notation="LaTeX">$36\times $ </tex-math></inline-formula> and <inline-formula> <tex-math notation="LaTeX">$33\times $ </tex-math></inline-formula> compared to Kraken2 and MetaCache (2 state-of-the-art profilers), respectively, on typical food-related databases. Demeter maintains an acceptable profiling accuracy (within 2% of existing tools) and incurs a very low area overhead. |
| first_indexed | 2024-04-12T06:33:21Z |
| format | Article |
| id | doaj.art-1d358a0450c84dbda5fc51e36f66f851 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-12T06:33:21Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-1d358a0450c84dbda5fc51e36f66f8512022-12-22T03:43:56ZengIEEEIEEE Access2169-35362022-01-0110824938251010.1109/ACCESS.2022.31958789847238Demeter: A Fast and Energy-Efficient Food Profiler Using Hyperdimensional Computing in MemoryTaha Shahroodi0https://orcid.org/0000-0003-4576-0030Mahdi Zahedi1https://orcid.org/0000-0002-7602-5066Can Firtina2Mohammed Alser3https://orcid.org/0000-0002-6117-3701Stephan Wong4Onur Mutlu5https://orcid.org/0000-0002-0075-2312Said Hamdioui6https://orcid.org/0000-0002-8961-0387Q&CE Department, EEMCS Faculty, Delft University of Technology (TU Delft), Delft, The NetherlandsQ&CE Department, EEMCS Faculty, Delft University of Technology (TU Delft), Delft, The NetherlandsSAFARI Research Group, D-ITET, ETH Zürich, Zürich, SwitzerlandSAFARI Research Group, D-ITET, ETH Zürich, Zürich, SwitzerlandQ&CE Department, EEMCS Faculty, Delft University of Technology (TU Delft), Delft, The NetherlandsSAFARI Research Group, D-ITET, ETH Zürich, Zürich, SwitzerlandQ&CE Department, EEMCS Faculty, Delft University of Technology (TU Delft), Delft, The NetherlandsFood profiling is an essential step in any food monitoring system needed to prevent health risks and potential frauds in the food industry. Significant improvements in sequencing technologies are pushing food profiling to become the main computational bottleneck. State-of-the-art profilers are unfortunately too costly for food profiling. Our goal is to design a food profiler that solves the main limitations of existing profilers, namely (1) working on massive data structures and (2) incurring considerable data movement, for a real-time monitoring system. To this end, we propose Demeter, the first platform-independent framework for food profiling. Demeter overcomes the first limitation through the use of hyperdimensional computing (HDC) and efficiently performs the accurate few-species classification required in food profiling. We overcome the second limitation by the use of an in-memory hardware accelerator for Demeter (named Acc-Demeter) based on memristor devices. Acc-Demeter actualizes several domain-specific optimizations and exploits the inherent characteristics of memristors to improve the overall performance and energy consumption of Acc-Demeter. We compare Demeter’s accuracy with other industrial food profilers using detailed software modeling. We synthesize Acc-Demeter’s required hardware using UMC’s 65nm library by considering an accurate PCM model based on silicon-based prototypes. Our evaluations demonstrate that Acc-Demeter achieves a (1) throughput improvement of <inline-formula> <tex-math notation="LaTeX">$192\times $ </tex-math></inline-formula> and <inline-formula> <tex-math notation="LaTeX">$724\times $ </tex-math></inline-formula> and (2) memory reduction of <inline-formula> <tex-math notation="LaTeX">$36\times $ </tex-math></inline-formula> and <inline-formula> <tex-math notation="LaTeX">$33\times $ </tex-math></inline-formula> compared to Kraken2 and MetaCache (2 state-of-the-art profilers), respectively, on typical food-related databases. Demeter maintains an acceptable profiling accuracy (within 2% of existing tools) and incurs a very low area overhead.https://ieeexplore.ieee.org/document/9847238/Food profilingemerging memoriesin memory processinganalog computing |
| spellingShingle | Taha Shahroodi Mahdi Zahedi Can Firtina Mohammed Alser Stephan Wong Onur Mutlu Said Hamdioui Demeter: A Fast and Energy-Efficient Food Profiler Using Hyperdimensional Computing in Memory IEEE Access Food profiling emerging memories in memory processing analog computing |
| title | Demeter: A Fast and Energy-Efficient Food Profiler Using Hyperdimensional Computing in Memory |
| title_full | Demeter: A Fast and Energy-Efficient Food Profiler Using Hyperdimensional Computing in Memory |
| title_fullStr | Demeter: A Fast and Energy-Efficient Food Profiler Using Hyperdimensional Computing in Memory |
| title_full_unstemmed | Demeter: A Fast and Energy-Efficient Food Profiler Using Hyperdimensional Computing in Memory |
| title_short | Demeter: A Fast and Energy-Efficient Food Profiler Using Hyperdimensional Computing in Memory |
| title_sort | demeter a fast and energy efficient food profiler using hyperdimensional computing in memory |
| topic | Food profiling emerging memories in memory processing analog computing |
| url | https://ieeexplore.ieee.org/document/9847238/ |
| work_keys_str_mv | AT tahashahroodi demeterafastandenergyefficientfoodprofilerusinghyperdimensionalcomputinginmemory AT mahdizahedi demeterafastandenergyefficientfoodprofilerusinghyperdimensionalcomputinginmemory AT canfirtina demeterafastandenergyefficientfoodprofilerusinghyperdimensionalcomputinginmemory AT mohammedalser demeterafastandenergyefficientfoodprofilerusinghyperdimensionalcomputinginmemory AT stephanwong demeterafastandenergyefficientfoodprofilerusinghyperdimensionalcomputinginmemory AT onurmutlu demeterafastandenergyefficientfoodprofilerusinghyperdimensionalcomputinginmemory AT saidhamdioui demeterafastandenergyefficientfoodprofilerusinghyperdimensionalcomputinginmemory |