FPGA Acceleration of Rigid Molecule Interactions

Abstract

Modeling of molecule interactions often uses two or more successive models of increasing complexity. Rigid models based on correlation techniques are common as early screening passes-to detect interactions worth costlier examination-and are often at the heart of later passes as well. Even these rigid models are time-consuming when applied to large models at 10 3 − 10 5 different three-axis rotations. This paper presents an FPGA structure for performing the correlations efficiently by using a systolic array for 3-D correlation and arithmetic tailored to the application. The system includes a novel addressing technique for performing a three-axis rotation of a 3-D voxel model using modest amounts of logic and nearly no cost in time or buffer space. We compare our FPGA implementation with one on a PC using the standard transform-based method and find a speed-up of a factor of 200. We present extensions for handling implementation technologies with different performance characteristics and for handling models too large to fit on-chip.

End-to-End Big Data Processing Protection in Cloud Environment Using Black Boxes - An Fpga Approach

“…There are recent efforts on applying FPGAs as accelerators for parallel data analytics (Court et al, 2004;Ronan et al, 2006;Woods & Van Court, 2008). Shan et.al present a framework to use FPGAs to accelerate MapReduce processing in (Shan et al, 2010).…”

“…FPGAs are considered to be powerful computation devices and suited as accelerators for b ig data analytics. It is common for many applications to employ FPGA based devices to accelerate their performance (e.g., (Court, Gu & Herbordt, 2004;Ronan, Éigeartaigh, Murphy, Scott & Kerins, 2006;Woods & VanCourt, 2008)). FPGAs are also deployed in the cloud environment as accelerators for large scale data processing such as MapReduce (Shan, Wang, Yan, Wang, Xu & Yang, 2010).…”

Acceleration of a production rigid molecule docking code

“…Non-intersection Poor fit Figure 2: Shape complementarity, collisions, misses, and poor matches (from [10]). …”

“…The introduction of the FFT to docking [6] reduced the complexity of each 3D correlation to Ç´AE ¿ ÐÓ AEµ for steric (shape only) models; further work expanded the method to electrostatic [3] and solvation contributions [1]. In previous work [10,11] we showed that, for FPGAbased coprocessors, the original direct correlationrather than an FFT-is sometimes the preferred method for computing rigid molecule docking. Two reasons for this are the inherent efficiency with which FPGAs perform convolutions and the modest precision (2-7 bits) of the original voxel data.…”

Accelerating molecular dynamics simulations with configurable circuits

“…Finally, this work is part of a larger project involving the acceleration of applications in computational biochemistry (e.g. [14]) and will be integrated into that.…”