Server-based execution of periodic tasks on dynamically reconfigurable hardware

Abstract

A prototype system that executes a set of periodic real-time tasks utilising dynamic hardware reconfiguration is presented. The proposed scheduling technique, merge server distribute load (MSDL), is not only able to give an offline guarantee for the feasibility of the task set, but also minimises the number of device configurations. After describing this technique, the schedulability analysis is extended to cover different runtime system overheads, including the device reconfiguration time. Then, a lightweight runtime system that performs the online part of the MSDL scheduling technique is detailed. The runtime system is implemented entirely in hardware. Finally, the corresponding synthesis tool flow is outlined and the overhead posed by the runtime system is reported.

ReconOS

“…This involves a supporting hardware infrastructure, which is already in place, as well as the investigation of suitable scheduling techniques. We plan to leverage our existing work on scheduling (see Danne et al [2006Danne et al [ , 2007) and extend the host operating system's schedulers to deal with loadable hardware threads. -Communication Primitives.…”

“…A prominent example is placement and scheduling of hardware tasks, which has been studied under a variety of task and resource models as well as optimization objectives. Examples can be found in Jean et al [1999], Bazargan et al [2000], Teich et al [2000], Steiger et al [2004], Danne and Platzner [2006], Danne et al [2007], and Pellizzoni and • 8:5 Caccamo [2007]. Many efficient scheduling techniques, especially for real-time systems, rely on task preemption.…”

Efficient Scheduling of FPGAs for Cloud Data Center Infrastructures

“…In [12], a server is defined as a periodic task whose purpose is to serve aperiodic requests for resources as soon as possible. Static server-based scheduling for FPGAs was first studied in [13] for independent periodic tasks. On-line server-based scheduling is described in [14], [15], in the context of a realtime operating system, also for independent tasks.…”

“…On-line server-based scheduling is described in [14], [15], in the context of a realtime operating system, also for independent tasks. Our work di ers from [13]- [15] as (i) we account for task dependencies and (ii) we provide a generic model for : resources whose requests are constant in time and independent of task scheduling.…”