{"id":1323,"date":"2021-05-10T11:55:58","date_gmt":"2021-05-10T09:55:58","guid":{"rendered":"http:\/\/emg2.com\/?page_id=1323"},"modified":"2022-04-04T17:41:12","modified_gmt":"2022-04-04T15:41:12","slug":"computing-hardware-acceleration","status":"publish","type":"page","link":"https:\/\/emg2.com\/en\/solutions\/computing-hardware-acceleration\/","title":{"rendered":"Computing & Hardware Acceleration"},"content":{"rendered":"

[vc_row css_animation=”” row_type=”row” use_row_as_full_screen_section=”no” type=”full_width” angled_section=”no” text_align=”left” background_image_as_pattern=”without_pattern” css=”.vc_custom_1608110893141{padding-top: 200px !important;padding-bottom: 200px !important;}” z_index=”” background_image=”10″][vc_column][\/vc_column][\/vc_row][vc_row css_animation=”” row_type=”row” use_row_as_full_screen_section=”no” type=”full_width” angled_section=”no” text_align=”left” background_image_as_pattern=”without_pattern”][vc_column][vc_row_inner row_type=”row” type=”grid” text_align=”left” css_animation=”” css=”.vc_custom_1608112591010{margin-bottom: 60px !important;}”][vc_column_inner][vc_column_text el_class=”title-page”]<\/p>\n

Computing & Hardware Acceleration<\/strong><\/h1>\n

[\/vc_column_text][\/vc_column_inner][\/vc_row_inner][vc_row_inner row_type=”row” type=”grid” text_align=”left” css_animation=”” el_class=”detail-row” css=”.vc_custom_1621932903989{margin-bottom: 60px !important;}” el_id=”FPGA_board”][vc_column_inner width=”2\/5″][vc_column_text el_class=”detail-subtitle”]<\/p>\n

FPGA Boards \/<\/b>\/<\/b>\/<\/b><\/h2>\n

[\/vc_column_text][vc_single_image image=”975″ img_size=”370×300″ qode_css_animation=””][\/vc_column_inner][vc_column_inner width=”3\/5″][vc_column_text css=”.vc_custom_1649086706497{padding-top: 30px !important;}”]<\/p>\n

Whether to accelerate<\/strong> or increase<\/strong> the performance of various algorithms<\/strong>, for very high-speed data transmission<\/strong> and real-time processing of all kinds of signals<\/strong>, our FPGA platforms are interesting alternatives to complement and even replace traditional processors<\/strong>. Available in industrial standards (PCIe, AMC, VPX)<\/strong> or more specific<\/strong> form factors, with various memory<\/strong> and interface options (FMC connectors, network links, NVMe, …)<\/strong>, these boards can be integrated into embedded systems<\/strong> as well as servers or industrial PCs<\/strong> to cover their numerous use cases.\n<\/p>\n

With the recent improvements of FPGA programming environments<\/strong> and the availability of functional examples<\/strong>, IPs and Frameworks<\/strong>, the exploitation of all the computing resources of our boards are simplified<\/strong> while guaranteeing the agility and scalability<\/strong> of your projects.<\/p>\n

[\/vc_column_text][vc_column_text el_class=”liste-application”]M<\/i><\/a>Y<\/i><\/a>U<\/i><\/a>X<\/i><\/a>S<\/i><\/a>j<\/i><\/a>E<\/i><\/a>I<\/i><\/a>[\/vc_column_text][\/vc_column_inner][\/vc_row_inner][vc_row_inner row_type=”row” type=”grid” text_align=”left” css_animation=”” el_class=”detail-row” css=”.vc_custom_1621932911798{margin-bottom: 60px !important;}” el_id=”AI”][vc_column_inner width=”3\/5″][vc_column_text el_class=”detail-subtitle”]<\/p>\n

Machine Learning \/ AI \/<\/b>\/<\/b>\/<\/b><\/h2>\n

[\/vc_column_text][vc_column_text css=”.vc_custom_1649086828310{padding-top: 30px !important;}”]<\/p>\n

When it comes to reducing power consumption and system footprint<\/strong> of processing nodes (Edge Computing<\/strong> platforms for example), our FPGA-based solutions become obvious, especially given their remarkable performances<\/strong> to accelerate some algorithms like inference tasks for artificial intelligence.<\/strong>
\nWe offer FPGA platforms dedicated to Machine Learning<\/strong>, both for embedded systems (Machine Vision<\/strong> for example) and datacenters<\/strong>, whether in the form of PCIe boards<\/strong> (similar to GPU boards generally used for such functions) or more compact modules (SSD form factor)<\/strong> to be integrated within the storage servers<\/strong> for Big Data Analysis and database acceleration<\/strong> applications for example.<\/p>\n

[\/vc_column_text][vc_column_text el_class=”liste-application”]M<\/i><\/a>Y<\/i><\/a>U<\/i><\/a>X<\/i><\/a>S<\/i><\/a>j<\/i><\/a>E<\/i><\/a>I<\/i><\/a>[\/vc_column_text][\/vc_column_inner][vc_column_inner width=”2\/5″][vc_single_image image=”801″ img_size=”370×300″ alignment=”right” qode_css_animation=””][\/vc_column_inner][\/vc_row_inner][vc_row_inner row_type=”row” type=”grid” text_align=”left” css_animation=”” el_class=”detail-row” css=”.vc_custom_1621932930793{margin-bottom: 60px !important;}” el_id=”computational_storage”][vc_column_inner width=”2\/5″][vc_column_text el_class=”detail-subtitle”]<\/p>\n

Computational Storage<\/b> \/<\/b>\/<\/b>\/<\/b><\/h2>\n

[\/vc_column_text][vc_single_image image=”802″ img_size=”370×300″ qode_css_animation=””][\/vc_column_inner][vc_column_inner width=”3\/5″][vc_column_text css=”.vc_custom_1649086841812{padding-top: 30px !important;}”]<\/p>\n

Traditional data storage architectures<\/strong> based on SSD hard drives interconnected on PCIe and NVMe busses imply numerous limitations<\/strong>, whether in terms of latency<\/strong>, data transfer efficiency<\/strong> or processor availability<\/strong>. Our FPGA modules in U.2 or M.2 format (SSD form factor) are integrated directly on the PCIe or NVMe bus<\/strong> to act as compute nodes closely tied to the data<\/strong> while ensuring CPU Offload.<\/strong>
\nIn addition to the possibilities offered by this type of architecture for database acceleration<\/strong>, it is above all an enormous gain on a more global scale<\/strong> with the possibility to perfom smart storage (Erasure Coding, Deduplication)<\/strong> as well as compression and encryption functions<\/strong> in a more efficient way by optimizing data transfers and consumption.<\/p>\n

[\/vc_column_text][vc_column_text el_class=”liste-application”]M<\/i><\/a>Y<\/i><\/a>U<\/i><\/a>X<\/i><\/a>S<\/i><\/a>j<\/i><\/a>E<\/i><\/a>I<\/i><\/a>[\/vc_column_text][\/vc_column_inner][\/vc_row_inner][vc_row_inner row_type=”row” type=”grid” text_align=”left” css_animation=”” el_class=”detail-row” css=”.vc_custom_1621932960486{margin-bottom: 60px !important;}” el_id=”FPGA_servers”][vc_column_inner width=”3\/5″][vc_column_text el_class=”detail-subtitle”]<\/p>\n

FPGA Servers and systems \/<\/b>\/<\/b>\/<\/b><\/h2>\n

[\/vc_column_text][vc_column_text css=”.vc_custom_1649086854416{padding-top: 30px !important;}”]<\/p>\n

To improve the TTM or simplify the use of our various FPGA solutions, we also offer pre-integrated solutions<\/strong>, including one or more FPGA boards interconnected and pre-configured<\/strong> around processors and various hardware and software resources.
\nAvailable in the form of servers (Teraboxes)<\/strong> or 19” chassis of various sizes (from 1U to 4U)<\/strong>, these solutions are ready to be integrated into computing racks or embedded systems for rapid implementation<\/strong>. The user just has to focus on the installation of his algorithms and applications<\/strong> to have a turnkey solution<\/strong> for numerous use cases.<\/p>\n

[\/vc_column_text][vc_column_text el_class=”liste-application”]M<\/i><\/a>Y<\/i><\/a>U<\/i><\/a>X<\/i><\/a>S<\/i><\/a>j<\/i><\/a>E<\/i><\/a>[\/vc_column_text][\/vc_column_inner][vc_column_inner width=”2\/5″][vc_single_image image=”689″ img_size=”370×300″ alignment=”right” qode_css_animation=””][\/vc_column_inner][\/vc_row_inner][\/vc_column][\/vc_row][vc_row css_animation=”” row_type=”row” use_row_as_full_screen_section=”no” type=”full_width” angled_section=”no” text_align=”left” background_image_as_pattern=”without_pattern” css=”.vc_custom_1610387197745{margin-bottom: 65px !important;}” z_index=”” el_class=”hidden”][vc_column][vc_row_inner row_type=”row” type=”grid” text_align=”left” css_animation=””][vc_column_inner][vc_column_text el_class=”detail-subtitle” css=”.vc_custom_1610120930299{margin-bottom: 40px !important;}”]<\/p>\n

Markets \/<\/b>\/<\/b>\/<\/b><\/h2>\n

[\/vc_column_text][\/vc_column_inner][\/vc_row_inner][vc_row_inner row_type=”row” type=”grid” text_align=”left” css_animation=””][vc_column_inner width=”1\/6″][vc_raw_html]JTNDaSUyMGNsYXNzJTNEJTIyaWNvbiUyMGljb24tYXBwbGljYXRpb25zJTIyJTNFJTI2JTIzeDAwNDMlM0IlM0MlMkZpJTNF[\/vc_raw_html][vc_column_text]<\/p>\n

Transport<\/p>\n

[\/vc_column_text][\/vc_column_inner][vc_column_inner width=”1\/6″][\/vc_column_inner][vc_column_inner width=”1\/6″][\/vc_column_inner][vc_column_inner width=”1\/6″][\/vc_column_inner][vc_column_inner width=”1\/6″][\/vc_column_inner][vc_column_inner width=”1\/6″][\/vc_column_inner][\/vc_row_inner][\/vc_column][\/vc_row][vc_row css_animation=”” row_type=”row” use_row_as_full_screen_section=”no” type=”full_width” angled_section=”no” text_align=”left” background_image_as_pattern=”without_pattern” css=”.vc_custom_1610387214147{margin-bottom: 50px !important;}” z_index=”” el_class=”hidden”][vc_column][vc_row_inner row_type=”row” type=”grid” text_align=”left” css_animation=””][vc_column_inner][vc_column_text el_class=”detail-subtitle” css=”.vc_custom_1610120985193{margin-bottom: 40px !important;}”]<\/p>\n

Partners \/<\/b>\/<\/b>\/<\/b><\/h2>\n

[\/vc_column_text][\/vc_column_inner][\/vc_row_inner][vc_row_inner row_type=”row” type=”grid” text_align=”left” css_animation=””][vc_column_inner]