Software portability

About: Software portability is a research topic. Over the lifetime, 8987 publications have been published within this topic receiving 164922 citations. The topic is also known as: portability.

Go Ahead: A Partial Reconfiguration Framework

Abstract: Exploiting the benefits of partial run-time reconfiguration requires efficient tools. In this paper, we introduce the tool Go Ahead that is able to implement run-time reconfigurable systems for all recent Xilinx FPGAs. This includes in particular support for low cost and low power Spartan-6 FPGAs. Go Ahead assists during floor planning and automates the constraint generation. It interacts with the Xilinx vendor tools and triggers the physical implementation phases all the way down to the final configuration bit streams. Go Ahead enables the building of flexible systems for integrating many reconfigurable modules very efficiently into a system. The tool targets (re)usability, portability to future devices, and migration paths among reconfigurable systems featuring different FPGAs or even FPGA families. Moreover, it provides a scripting interface and all features can be accessed remotely.

From single-user architectural design to PAC*: a generic software architecture model for CSCW

Abstract: This article nqmts our tiection on softsvam amhiteetm modelling for multi-user systems (or groupware).Fi~ w introduce t.k notion of software architecture and make explieit the design steps that most sofl-wam designm in HCI tend to blend in a fizzy way. Building on general concepts and pmetiee fmm main stream soikm engineering, we then present a comparative analysis of the most signifhnt architecture models developed for singleand multi-user systems. We close with the pmentation d PAC*, a new arehitectuml fmmwodc for modelling and designing the sdsvare amhitectm of multi-user systems. PAC* is a motivated combination of existing anAiteetuml models seleeted for the complementarily of their “good properties”. These include operational heuristics such as rules for deriving agents in accordance to the task model or criteria for reasoning about replication as well as properties such as support for style heterogeneity, portability, and reusability.

TVM: An Automated End-to-End Optimizing Compiler for Deep Learning

Abstract: There is an increasing need to bring machine learning to a wide diversity of hardware devices. Current frameworks rely on vendor-specific operator libraries and optimize for a narrow range of server-class GPUs. Deploying workloads to new platforms -- such as mobile phones, embedded devices, and accelerators (e.g., FPGAs, ASICs) -- requires significant manual effort. We propose TVM, a compiler that exposes graph-level and operator-level optimizations to provide performance portability to deep learning workloads across diverse hardware back-ends. TVM solves optimization challenges specific to deep learning, such as high-level operator fusion, mapping to arbitrary hardware primitives, and memory latency hiding. It also automates optimization of low-level programs to hardware characteristics by employing a novel, learning-based cost modeling method for rapid exploration of code optimizations. Experimental results show that TVM delivers performance across hardware back-ends that are competitive with state-of-the-art, hand-tuned libraries for low-power CPU, mobile GPU, and server-class GPUs. We also demonstrate TVM's ability to target new accelerator back-ends, such as the FPGA-based generic deep learning accelerator. The system is open sourced and in production use inside several major companies.

Portability regimes of pension and health care benefits for international migrants: an analysis of issues and good practices

Abstract: The paper provides a first investigation into the portability of pension and health care benefits for international migrants. It is based on available literature and newly minted data, but more importantly on selective case studies from main migrant-sending and receiving countries. While exploratory, the paper achieves a better understanding of the realities on the ground and is able to distill key issues as well as identify good and best practices. The main conclusions include the following: First, only around 20 percent of migrants worldwide work in host countries where full portability of pension benefits, but not necessarily of health care benefits, to their home countries is ensured. Second, bilateral agreements are seemingly the current best practice to ensure portability for pension and health care benefits, although for the latter this is not always the case. Third, more actuarial-type structures should help to enhance portability. This is, in principle, straightforward for pensions and a defined contribution-type design. It is much more complicated for health care benefits. Last but not least, for improved benefit design and implementation, the information base needs to be broadened, including through more country case studies and tracer studies of migrants.