We propose a new method for defragmenting the module layout of a reconfigurable device, enabled by a novel approach for dealing with communication needs between relocated modules and with inhomogeneities found in commonly used FPGAs. Our method is based on dynamic relocation of module positions during runtime, with only very little reconfiguration overhead; the objective is to maximize the length of contiguous free space that is available for new modules. We describe a number of algorithmic aspects of good defragmentation, and present an optimization method based on tabu search. Experimental results indicate that we can improve the quality of module layout by roughly 50% over static layout. Among other benefits, this improvement avoids unnecessary rejections of modules. * A preliminary, considerably shorter extended-abstract version of this paper appeared in the proceedings of the FPL 08 [8].

No abstract

We propose a new method for defragmenting the module layout of a reconfigurable device, enabled by a novel approach for dealing with communication needs between relocated modules and with inhomogeneities found in commonly used FPGAs. Our method is based on dynamic relocation of module positions during runtime, with only very little reconfiguration overhead; the objective is to maximize the length of contiguous free space that is available for new modules. We describe a number of algorithmic aspects of good defragmentation, and present an optimization method based on tabu search. Experimental results indicate that we can improve the quality of module layout by roughly 50% over static layout. Among other benefits, this improvement avoids unnecessary rejection of modules.

Abstract. In this paper we consider methods for dynamically storing a set of different objects ("modules") in a physical array. Each module requires one free contiguous subinterval in order to be placed. Items are inserted or removed, resulting in a fragmented layout that makes it harder to insert further modules. It is possible to relocate modules, one at a time, to another free subinterval that is contiguous and does not overlap with the current location of the module. These constraints clearly distinguish our problem from classical memory allocation. We present a number of algorithmic results, including a bound of Θ(n 2 ) on physical sorting if there is a sufficiently large free space and sum up NP-hardness results for arbitrary initial layouts. For online scenarios in which modules arrive one at a time, we present a method that requires O(1) moves per insertion or deletion and amortized cost O(mi lgm) per insertion or deletion, where mi is the module's size,m is the size of the largest module and costs for moves are linear in the size of a module.

We consider the following online allocation problem: Given a unit square S, and a sequence of numbers n i ∈ {0, 1} with i j=0 n j ≤ 1; at each step i, select a region C i of previously unassigned area n i in S. The objective is to make these regions compact in a distance-aware sense: minimize the maximum (normalized) average Manhattan distance between points from the same set C i . Related location problems have received a considerable amount of attention; in particular, the problem of determining the "optimal shape of a city", i.e., allocating a single n i has been studied. We present an online strategy, based on an analysis of space-filling curves; for continuous shapes, we prove a factor of 1.8092, and 1.7848 for discrete point sets.