arXiv subjects

Peter Sanders

Publications and source records attributed to Peter Sanders.

125 records · Page 7Linked to original sources

arXiv2013

Transit Node Routing Reconsidered

Transit Node Routing (TNR) is a fast and exact distance oracle for road networks. We show several new results for TNR. First, we give a surprisingly simple implementation fully based on Contraction Hierarchies that speeds up preprocessing by an order of magnitude approaching the time for just finding a CH (which alone has two orders of magnitude larger query time). We also develop a very effective purely graph theoretical locality filter without any compromise in query times. Finally, we show that a specialization to the online many-to-one (or one-to-many) shortest path further speeds up query time by an order of magnitude. This variant even has better query time than the fastest known previous methods which need much more space.

cs.DS↗

arXiv2012

Think Locally, Act Globally: Perfectly Balanced Graph Partitioning

We present a novel local improvement scheme for the perfectly balanced graph partitioning problem. This scheme encodes local searches that are not restricted to a balance constraint into a model allowing us to find combinations of these searches maintaining balance by applying a negative cycle detection algorithm. We combine this technique with an algorithm to balance unbalanced solutions and integrate it into a parallel multi-level evolutionary algorithm, KaFFPaE, to tackle the problem. Overall, we obtain a system that is fast on the one hand and on the other hand is able to improve or reproduce most of the best known perfectly balanced partitioning results ever reported in the literature.

cs.DS↗

arXiv2012

Advanced Coarsening Schemes for Graph Partitioning

The graph partitioning problem is widely used and studied in many practical and theoretical applications. The multilevel strategies represent today one of the most effective and efficient generic frameworks for solving this problem on large-scale graphs. Most of the attention in designing the multilevel partitioning frameworks has been on the refinement phase. In this work we focus on the coarsening phase, which is responsible for creating structurally similar to the original but smaller graphs. We compare different matching- and AMG-based coarsening schemes, experiment with the algebraic distance between nodes, and demonstrate computational results on several classes of graphs that emphasize the running time and quality advantages of different coarsenings.

cs.DS↗

arXiv2011

Distributed Evolutionary Graph Partitioning

We present a novel distributed evolutionary algorithm, KaFFPaE, to solve the Graph Partitioning Problem, which makes use of KaFFPa (Karlsruhe Fast Flow Partitioner). The use of our multilevel graph partitioner KaFFPa provides new effective crossover and mutation operators. By combining these with a scalable communication protocol we obtain a system that is able to improve the best known partitioning results for many inputs in a very short amount of time. For example, in Walshaw's well known benchmark tables we are able to improve or recompute 76% of entries for the tables with 1%, 3% and 5% imbalance.

cs.NE↗

arXiv2011

Engineering Multilevel Graph Partitioning Algorithms

We present a multi-level graph partitioning algorithm using novel local improvement algorithms and global search strategies transferred from the multi-grid community. Local improvement algorithms are based max-flow min-cut computations and more localized FM searches. By combining these techniques, we obtain an algorithm that is fast on the one hand and on the other hand is able to improve the best known partitioning results for many inputs. For example, in Walshaw's well known benchmark tables we achieve 317 improvements for the tables 1%, 3% and 5% imbalance. Moreover, in 118 additional cases we have been able to reproduce the best cut in this benchmark.

cs.DS↗

arXiv2011

Efficient Error-Correcting Geocoding

We study the problem of resolving a perhaps misspelled address of a location into geographic coordinates of latitude and longitude. Our data structure solves this problem within a few milliseconds even for misspelled and fragmentary queries. Compared to major geographic search engines such as Google or Bing we achieve results of significantly better quality.

cs.IR↗

arXiv2010

Compressed Transmission of Route Descriptions

We present two methods to compress the description of a route in a road network, i.e., of a path in a directed graph. The first method represents a path by a sequence of via edges. The subpaths between the via edges have to be unique shortest paths. Instead of via edges also via nodes can be used, though this requires some simple preprocessing. The second method uses contraction hierarchies to replace subpaths of the original path by shortcuts. The two methods can be combined with each other. Also, we propose the application to mobile server based routing: We compute the route on a server which has access to the latest information about congestions for example. Then we transmit the computed route to the car using some mobile radio communication. There, we apply the compression to save costs and transmission time. If the compression works well, we can transmit routes even when the bandwidth is low. Although we have not evaluated our ideas with realistic data yet, they are quite promising.

cs.DS↗

arXiv2010

Faster Radix Sort via Virtual Memory and Write-Combining

Sorting algorithms are the deciding factor for the performance of common operations such as removal of duplicates or database sort-merge joins. This work focuses on 32-bit integer keys, optionally paired with a 32-bit value. We present a fast radix sorting algorithm that builds upon a microarchitecture-aware variant of counting sort. Taking advantage of virtual memory and making use of write-combining yields a per-pass throughput corresponding to at least 88 % of the system's peak memory bandwidth. Our implementation outperforms Intel's recently published radix sort by a factor of 1.5. It also compares favorably to the reported performance of an algorithm for Fermi GPUs when data-transfer overhead is included. These results indicate that scalar, bandwidth-sensitive sorting algorithms remain competitive on current architectures. Various other memory-intensive applications can benefit from the techniques described herein.

cs.DS↗

arXiv2010

Improved Fast Similarity Search in Dictionaries

We engineer an algorithm to solve the approximate dictionary matching problem. Given a list of words $\mathcal{W}$, maximum distance $d$ fixed at preprocessing time and a query word $q$, we would like to retrieve all words from $\mathcal{W}$ that can be transformed into $q$ with $d$ or less edit operations. We present data structures that support fault tolerant queries by generating an index. On top of that, we present a generalization of the method that eases memory consumption and preprocessing time significantly. At the same time, running times of queries are virtually unaffected. We are able to match in lists of hundreds of thousands of words and beyond within microseconds for reasonable distances.

cs.IR↗

arXiv2010

n-Level Graph Partitioning

We present a multi-level graph partitioning algorithm based on the extreme idea to contract only a single edge on each level of the hierarchy. This obviates the need for a matching algorithm and promises very good partitioning quality since there are very few changes between two levels. Using an efficient data structure and new flexible ways to break local search improvements early, we obtain an algorithm that scales to large inputs and produces the best known partitioning results for many inputs. For example, in Walshaw's well known benchmark tables we achieve 155 improvements dominating the entries for large graphs.

cs.DS↗

arXiv2010

Engineering a Scalable High Quality Graph Partitioner

We describe an approach to parallel graph partitioning that scales to hundreds of processors and produces a high solution quality. For example, for many instances from Walshaw's benchmark collection we improve the best known partitioning. We use the well known framework of multi-level graph partitioning. All components are implemented by scalable parallel algorithms. Quality improvements compared to previous systems are due to better prioritization of edges to be contracted, better approximation algorithms for identifying matchings, better local search heuristics, and perhaps most notably, a parallelization of the FM local search algorithm that works more locally than previous approaches.

cs.DC↗

arXiv2010

Defining and Computing Alternative Routes in Road Networks

Every human likes choices. But today's fast route planning algorithms usually compute just a single route between source and target. There are beginnings to compute alternative routes, but this topic has not been studied thoroughly. Often, the aspect of meaningful alternative routes is neglected from a human point of view. We fill in this gap by suggesting mathematical definitions for such routes. As a second contribution we propose heuristics to compute them, as this is NP-hard in general.

cs.DS↗

arXiv2009

Scalable Distributed-Memory External Sorting

We engineer algorithms for sorting huge data sets on massively parallel machines. The algorithms are based on the multiway merging paradigm. We first outline an algorithm whose I/O requirement is close to a lower bound. Thus, in contrast to naive implementations of multiway merging and all other approaches known to us, the algorithm works with just two passes over the data even for the largest conceivable inputs. A second algorithm reduces communication overhead and uses more conventional specifications of the result at the cost of slightly increased I/O requirements. An implementation wins the well known sorting benchmark in several categories and by a large margin over its competitors.

cs.DS↗

arXiv2009

GPU sample sort

In this paper, we present the design of a sample sort algorithm for manycore GPUs. Despite being one of the most efficient comparison-based sorting algorithms for distributed memory architectures its performance on GPUs was previously unknown. For uniformly distributed keys our sample sort is at least 25% and on average 68% faster than the best comparison-based sorting algorithm, GPU Thrust merge sort, and on average more than 2 times faster than GPU quicksort. Moreover, for 64-bit integer keys it is at least 63% and on average 2 times faster than the highly optimized GPU Thrust radix sort that directly manipulates the binary representation of keys. Our implementation is robust to different distributions and entropy levels of keys and scales almost linearly with the input size. These results indicate that multi-way techniques in general and sample sort in particular achieve substantially better performance than two-way merge sort and quicksort.

cs.DS↗

arXiv2009

Fast Detour Computation for Ride Sharing

Todays ride sharing services still mimic a better billboard. They list the offers and allow to search for the source and target city, sometimes enriched with radial search. So finding a connection between big cities is quite easy. These places are on a list of designated origin and distination points. But when you want to go from a small town to another small town, even when they are next to a freeway, you run into problems. You can't find offers that would or could pass by the town easily with little or no detour. We solve this interesting problem by presenting a fast algorithm that computes the offers with the smallest detours w.r.t. a request. Our experiments show that the problem is efficiently solvable in times suitable for a web service implementation. For realistic database size we achieve lookup times of about 5ms and a matching rate of 90% instead of just 70% for the simple matching algorithms used today.

cs.DS↗

arXiv2009

Towards Optimal Range Medians

We consider the following problem: given an unsorted array of $n$ elements, and a sequence of intervals in the array, compute the median in each of the subarrays defined by the intervals. We describe a simple algorithm which uses O(n) space and needs $O(n\log k + k\log n)$ time to answer the first $k$ queries. This improves previous algorithms by a logarithmic factor and matches a lower bound for $k=O(n)$. Since the algorithm decomposes the range of element values rather than the array, it has natural generalizations to higher dimensional problems -- it reduces a range median query to a logarithmic number of range counting queries.

cs.DS↗

arXiv2008

Time Dependent Contraction Hierarchies -- Basic Algorithmic Ideas

Contraction hierarchies are a simple hierarchical routing technique that has proved extremely efficient for static road networks. We explain how to generalize them to networks with time-dependent edge weights. This is the first hierarchical speedup technique for time-dependent routing that allows bidirectional query algorithms.

cs.DS↗