New-Tech Magazine - Europe | January Digital edition

A Novel Approach to Software-Defined FPGA Computing Stephane Monboisset , QuickPlay

W

ith the rise of the Internet of Things and Big Data

widespread FPGA adoption has been the complexity of implementing them. Until now, the only way to develop an application on an FPGA-based platform has been to deal with some of the lowest levels of hardware implementation. This has kept a large potential customer base-software developers-away from the devices and has made life increasingly complicated for traditional FPGA designers. Recent methodologies for FPGA design, centered on high-level synthesis (HLS) tools and leveraging software programming languages such as OpenCL™, C and C++, have provided a sandbox for software developers to reap the benefits of FPGA-based hardware acceleration in numerous applications. But the methodologies often fall short in one essential respect: enabling software developers to define and configure, on their own, the hardware infrastructure

best suited for their application. The industry has continued to pursue the holy grail of a high-level workflow for implementing applications on FPGA- based platforms that does not require specific FPGA expertise. Over the past five years, PLDA has developed just such a workflow. Called QuickPlay, it efficiently addresses the implementation complexity challenge and enables multiple use models for FPGA development. But one of its core sources of value is the way in which it lets software developers take applications intended for CPUs and implement them, partially or fully, on FPGA hardware. QuickPlay leverages all of the FPGA resources, turning these powerful but complex devices into software-defined platforms that yield the benefits of FPGAs without the pain of hardware design. Consider a software algorithm that can be broken down into two functions:

processing, the need for transferring and processing data has skyrocketed, and CPUs alone can no longer address the exponential increase. Adding more processors and more virtual machines to run a given application just doesn’t cut it, as there is only so much that can be parallelized on multiple CPUs for a given application. Field-programmable gate arrays, on the other hand, have the requisite I/O bandwidth and processing power, not only from a pure processing standpoint but, equally important, from a power standpoint. For data- center equipment manufacturers, the use of FPGAs has long been an appealing prospect. Intel’s recent acquisition of the second-largest FPGA vendor is further testament that a CPU-only solution no longer suffices. The major roadblock to more-

50 l New-Tech Magazine Europe