Kirill Minkovich

HRLSim: A High Performance Spiking Neural Network Simulator for GPGPU Clusters

Modeling of large-scale spiking neural models is an important tool in the quest to understand brain function and subsequently create real-world applications. This paper describes a spiking neural network simulator environment called HRL Spiking Simulator (HRLSim). This simulator is suitable for implementation on a cluster of general purpose graphical processing units (GPGPUs). Novel aspects of HRLSim are described and an analysis of its performance is provided for various configurations of the cluster. With the advent of inexpensive GPGPU cards and compute power, HRLSim offers an affordable and scalable tool for design, real-time simulation, and analysis of large-scale spiking neural networks.

Programming Time-Multiplexed Reconfigurable Hardware Using a Scalable Neuromorphic Compiler

Scalability and connectivity are two key challenges in designing neuromorphic hardware that can match biological levels. In this paper, we describe a neuromorphic system architecture design that addresses an approach to meet these challenges using traditional complementary metal-oxide-semiconductor (CMOS) hardware. A key requirement in realizing such neural architectures in hardware is the ability to automatically configure the hardware to emulate any neural architecture or model. The focus for this paper is to describe the details of such a programmable front-end. This programmable front-end is composed of a neuromorphic compiler and a digital memory, and is designed based on the concept of synaptic time-multiplexing (STM). The neuromorphic compiler automatically translates any given neural architecture to hardware switch states and these states are stored in digital memory to enable desired neural architectures. STM enables our proposed architecture to address scalability and connectivity using traditional CMOS hardware. We describe the details of the proposed design and the programmable front-end, and provide examples to illustrate its capabilities. We also provide perspectives for future extensions and potential applications.

5PM: Secure pattern matching

In this paper we consider the problem of secure pattern matching that allows single-character wildcards and substring matching in the malicious stand-alone setting. Our protocol, called 5PM, is executed between two parties: Server, holding a text of length n, and Client, holding a pattern of length m to be matched against the text, where our notion of matching is more general than traditionally considered and includes non-binary alphabets, non-binary Hamming distance and non-binary substring matching.5PM is the first secure expressive pattern matching protocol designed to optimize round complexity by carefully specifying the entire protocol round by round. 5PM requires only eight rounds in the malicious static corruptions model. In the malicious model, 5PM requires O((m+n)k2) communication complexity and O(m+n) encryptions, where m is the pattern length and n is the text length. Further, 5PM can hide pattern size with no asymptotic additional costs in either computation or bandwidth.

5PM: secure pattern matching

In this paper we consider the problem of secure pattern matching that allows single character wildcards and substring matching in the malicious (stand-alone) setting. Our protocol, called 5PM, is executed between two parties: Server, holding a text of length n, and Client, holding a pattern of length m to be matched against the text, where our notion of matching is more general and includes non-binary alphabets, non-binary Hamming distance and non-binary substring matching. 5PM is the first protocol with communication complexity sub-linear in circuit size to compute non-binary substring matching in the malicious model (general MPC has communication complexity which is at least linear in the circuit size). 5PM is also the first sublinear protocol to compute non-binary Hamming distance in the malicious model. Additionally, in the honest-but-curious (semi-honest) model, 5PM is asymptotically more efficient than the best known scheme when amortized for applications that require single charcter wildcards or substring pattern matching. 5PM in the malicious model requires O((m+n)k2) bandwidth and O(m+n) encryptions, where m is the pattern length and n is the text length. Further, 5PM can hide pattern size with no asymptotic additional costs in either computation or bandwidth. Finally, 5PM requires only 2 rounds of communication in the honest-but-curious model and 8 rounds in the malicious model. Our techniques reduce pattern matching and generalized Hamming distance problem to a novel linear algebra formulation that allows for generic solutions based on any additively homomorphic encryption. We believe our efficient algebraic techniques are of independent interest.

Efficiently passing messages in distributed spiking neural network simulation

Efficiently passing spiking messages in a neural model is an important aspect of high-performance simulation. As the scale of networks has increased so has the size of the computing systems required to simulate them. In addition, the information exchange of these resources has become more of an impediment to performance. In this paper we explore spike message passing using different mechanisms provided by the Message Passing Interface (MPI). A specific implementation, MVAPICH, designed for high-performance clusters with Infiniband hardware is employed. The focus is on providing information about these mechanisms for users of commodity high-performance spiking simulators. In addition, a novel hybrid method for spike exchange was implemented and benchmarked.