High latency in an interactive system limits its usability. In order to reduce end-to-end latency of such systems, it is necessary to analyze and optimize the latency of individual contributors, such as input devices, applications, or displays. We present a simple tool for measuring the latency of USB-connected input devices with sub-millisecond accuracy. The tool, based on a Raspberry Pi 2 mic...

background: herpes virus group can persist in the latent form in the body, pneumocystis jerovecii, on the other hand, is a fungus which is ubiquitous in the respiratory tract of mammals, with 90% of adult human beings possessing antibodies to this organism. concomitant presence of different microorganisms, may result in one microorganism altering the activity of another. objectives: to determin...

We propose an efficient performance-driven twoway partitioning algorithm to take into account clock period and latency with retiming. We model the problem with a Quadratic Programming formulation to minimize the crossing edge count with nonlinear timing constraints. By using a Lagrangian Approach on Modular Partitioning (LAMP), we merge nonlinear constraints into the objective function. We then...

In this paper we introduce a very low latency implementation of the Berkeley Sockets interface on top of the Dolphin SCI interconnect. Average one byte latencies as low as 2.31 microseconds are achieved. To our knowledge, this is the fastest sockets implementation in terms of latency. SCI SOCKET combines outstanding performance and seamless integration with legacy applications. This enables net...

Non-Crossing Ordered BDDs (NCOBDDs) [3], were proposed for mapping to pass transistor logic (PTL) circuits with very high performance and regular layout. However, the latency and area penalty prevents it to be applied to large circuits. In this paper, we propose a dynamic-programming based approach for decomposition of BDDs. The decomposed BDDs can be applied to construct more compact NCOBDDs t...

Advances in communication rates exceed the ability of a single source to fully utilize a gigabit WAN channel with existing deterministic protocols. Parallel communication describes a method for reducing latency by managing indeterminism and increasing channel utilization, given a surplus bandwidth-delay product. It involves a nondeterministic state mechanism, with a modified protocol interface....

To meet the performance and scalability demands of the fast-paced technological growth towards exascale and big data processing with the performance bottleneck of conventional metal-based interconnects (wireline), alternative interconnect fabrics, such as inhomogeneous three-dimensional integrated network-on-chip (3D NoC) and hybrid wired-wireless network-on-chip (WiNoC), have emanated as a cos...

In Disruption Tolerant Networks, packet forwarding scheme has a major drawback in terms of network congestion. In order to solve this problem, a New Adaptive Routing Proposal (NARP) has been proposed NARP uses the message forwarding probability scheme in order to increase the delivery ratio and bandwidth utilization. Also, NARP arranges the dropping sequence based on their assigned priority. Th...

We consider the automatic determination of application-specific memory subsystems via superoptimization, with the goals of reducing memory access time and of minimizing writes. The latter goal is of concern for memories with limited write endurance. Our subsystems outperform general-purpose memory subsystems in terms of performance, number of writes, or both.