Assisted Model Building with Energy Refinement* (Amber14*)

A suite of programs that allow users to carry out molecular dynamics simulations.

A Demonstrated 2x Performance Improvement1

Intel® software’s engineers and Professor Ross Walker of the Intel® Parallel Computing Center(s) (Intel® PCC) at the San Diego Supercomputer Center (SDSC) lead work on advanced manycore and multicore optimization of classical all atom molecular dynamics simulations within the life sciences arena. The work focuses on improving the performance of the Amber* molecular dynamics software on platforms based on Intel® Xeon® processors and Intel® Xeon Phi™ coprocessors for both single- and multi-node configurations.

Work conducted as part of the Intel® Parallel Computing Center(s) (Intel® PCC) feeds directly back into the Amber code base and also to the molecular dynamics community as a whole, impacting fields including drug discovery, biomaterials, and biocatalysis. Amber software is used in all major U.S. and international research universities as well as broadly across the pharmaceutical and biotech industries. Optimization and porting work for this project will thus directly benefit a large number of scientists.

Optimization Details

The baseline is the Intel® Xeon® processor E5-2697 v2 compared to the Intel® Xeon® processor E5-2697 v2 and Intel® Xeon Phi™ coprocessor 7120A, with offload processing on both and using the released code (double precision code across the platforms).

Performance Results

Optimized Intel® Xeon® processor E5-2697 v2 and Intel® Xeon Phi™ coprocessor 7120A offload demonstrated up to 2X improved performance over the baseline Intel® Xeon® processor E5-2697-only code.

Download the code ›

Reproduce these results with this optimization recipe (see section 18.7) ›

Configuration Table

System Overview Intel® Xeon® Processor E5-2697 v2 with Intel® Xeon Phi™ Coprocessor 7120A
Platform Intel® Server System R2208GZ4GC, 2U chassis, hot-swap drives, 24 DIMMs, 1 750W redundant power supply
CPU/Stepping Intel® Server System R2208GZ4GC, 2U chassis, hot-swap drives, 24 DIMMs, 1 750W redundant power supply
Coprocessor Intel® Xeon® processor E5-2697 v2, 2.7 GHz, 12 core, 8 GT/s dual Intel® QuickPath Interconnect (Intel® QPI) links, 130W, 3.5 GHz max turbo frequency, 768 kB instr L1 / 3072 kB L2 / 30 MB L3 cache
Memory Hynix HMT31GR7BFR-PB*, 64 GB total 8*8 GB 1600 MHZ Reg ECC DDR3
Chipset Rev 4.6, SE5C600.86B.99.99.x069.071520130923
BIOS BIOS configuration: Default except: Intel® Turbo Boost Technology enabled, Enhanced Intel SpeedStep® technology enabled, surface-mount technology (SMT) enabled, non-uniform memory access (NUMA) enabled, memory speed 1600 MHz, Note: All prefetchers enabled (this is default)
GigE Node Adapter Intel® Ethernet Controller I350 (rev 01), 4 Gigabit network connections, only one connection in use
IB Switch
36-port switch/rack Mellanox FDR* (model MSX6025F-1BFR), Firmware version: 9_2_4002
IB Adapters 36-port switch/rack Mellanox FDR (model MSX6025F-1BFR), Firmware version: 9_2_4002
HDD Specs Seagate ST9600205SS* (scsi), 1x600 GB SAS HDD 10kRPM
Operating System Red Hat Enterprise Linux* 6.4

Product and Performance Information

1

Benchmark results were obtained prior to implementation of recent software patches and firmware updates intended to address exploits referred to as "Spectre" and "Meltdown". Implementation of these updates may make these results inapplicable to your device or system.

Software and workloads used in performance tests may have been optimized for performance only on Intel® microprocessors. Performance tests, such as SYSmark* and MobileMark*, are measured using specific computer systems, components, software, operations, and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit http://www.intel.com/benchmarks.