Search

Fault-tolerant distributed applications require mechanisms to recover data lost via a process failure. On modern cluster systems it is typically impractical to request replacement resources after such a failure. Therefore, applications have to continue working with the remaining resources. This requires redistributing the workload and that the non-failed processes reload data. We present an algorithmic framework and its C++ library implementation ReStore for MPI programs that enables recovery of data after process failures. By storing all required data in memory via an appropriate data distribution and replication, recovery is substantially faster than with standard checkpointing schemes that rely on a parallel file system. As the application developer can specify which data to load, we also support shrinking recovery instead of recovery using spare compute nodes. We evaluate ReStore in both controlled, isolated environments and real applications. Our experiments show loading times of lost input data in the range of milliseconds on up to 24 576 processors and a substantial speedup of the recovery time for the fault-tolerant version of a widely used bioinformatics application.

Finding a maximum independent set is a fundamental NP-hard problem that is used in many real-world applications. Given an unweighted graph, this problem asks for a maximum cardinality set of pairwise non-adjacent vertices. In recent years, some of the most successful algorithms for solving this problem are based on the branch-and-bound or branch-and-reduce paradigms. In particular, branch-and-reduce algorithms, which combine branch-and-bound with reduction rules, have been able to achieve substantial results, solving many previously infeasible real-world instances. These results were to a large part achieved by developing new, more practical reduction rules. However, other components that have been shown to have a significant impact on the performance of these algorithms have not received as much attention. One of these is the branching strategy, which determines what vertex is included or excluded in a potential solution. Even now, the most commonly used strategy selects vertices solely based on their degree and does not take into account other factors that contribute to the performance of the algorithm. In this work, we develop and evaluate several novel branching strategies for both branch-and-bound and branch-and-reduce algorithms. Our strategies are based on one of two approaches which are motivated by existing research. They either (1) aim to decompose the graph into two or more connected components which can then be solved independently, or (2) try to remove vertices that hinder the application of a reduction rule which can lead to smaller graphs. Our experimental evaluation on a large set of real-world instances indicates that our strategies are able to improve the performance of the state-of-the-art branch-and-reduce algorithm by Akiba and Iwata. To be more specific, our reduction-based packing branching rule is able to outperform the default branching strategy of selecting a vertex of highest degree on 65% of all instances tested. Furthermore, our decomposition-based strategy based on edge cuts is able to achieve a speedup of 2.29 on sparse networks (1.22 on all instances)., LIPIcs, Vol. 190, 19th International Symposium on Experimental Algorithms (SEA 2021), pages 17:1-17:21

A highly successful approach to route planning in networks (particularly road networks) is to identify a hierarchy in the network that allows faster queries after some preprocessing that basically inserts additional "shortcut"-edges into a graph. In the past there has been a succession of techniques that infer a more and more fine grained hierarchy enabling increasingly more efficient queries. This appeared to culminate in contraction hierarchies that assign one hierarchy level to each vertex. In this paper we show how to identify an even more fine grained hierarchy that assigns one level to each edge of the network. Our findings indicate that this can lead to considerably smaller search spaces in terms of visited edges. Currently, this rarely implies improved query times so that it remains an open question whether edge hierarchies can lead to consistently improved performance. However, we believe that the technique as such is a noteworthy enrichment of the portfolio of available techniques that might prove useful in the future.

Kernelization is a general theoretical framework for preprocessing instances of NP-hard problems into (generally smaller) instances with bounded size, via the repeated application of data reduction rules. For the fundamental Max Cut problem, kernelization algorithms are theoretically highly efficient for various parameterizations. However, the efficacy of these reduction rules in practice---to aid solving highly challenging benchmark instances to optimality---remains entirely unexplored. We engineer a new suite of efficient data reduction rules that subsume most of the previously published rules, and demonstrate their significant impact on benchmark data sets, including synthetic instances, and data sets from the VLSI and image segmentation application domains. Our experiments reveal that current state-of-the-art solvers can be sped up by up to multiple orders of magnitude when combined with our data reduction rules. On social and biological networks in particular, kernelization enables us to solve four instances that were previously unsolved in a ten-hour time limit with state-of-the-art solvers; three of these instances are now solved in less than two seconds., 16 pages, 4 tables, 2 figures

We present the winning solver of the PACE 2019 Implementation Challenge, Vertex Cover Track. The minimum vertex cover problem is one of a handful of problems for which kernelization---the repeated reducing of the input size via data reduction rules---is known to be highly effective in practice. Our algorithm uses a portfolio of techniques, including an aggressive kernelization strategy, local search, branch-and-reduce, and a state-of-the-art branch-and-bound solver. Of particular interest is that several of our techniques were not from the literature on the vertex over problem: they were originally published to solve the (complementary) maximum independent set and maximum clique problems. Aside from illustrating our solver's performance in the PACE 2019 Implementation Challenge, our experiments provide several key insights not yet seen before in the literature. First, kernelization can boost the performance of branch-and-bound clique solvers enough to outperform branch-and-reduce solvers. Second, local search can significantly boost the performance of branch-and-reduce solvers. And finally, somewhat surprisingly, kernelization can sometimes make branch-and-bound algorithms perform worse than running branch-and-bound alone.

As a result of the growing amounts of Data in todays Databases, one machine is often not sufficient to store and process these. The proper solution to this problem is to scale the system out on a cluster. However, the distribution of the data throughout the machines of the cluster results in a high percentage of communication time in the overall execution time of a query, especially for complex analytical queries. For this reason, we try to minimize the volume of communicated data to allow faster runtimes when a query cannot be executed on a single node of the cluster without any communication. We analyze techniques from previous work and propose improvements to them backed by a complexity analysis of the communication volume for both, our algorithms and the algorithms from the previous work. For the evaluation of our algorithms we implement them for chosen queries of the TPC-H benchmark and run them on a cluster of up to 128 nodes with a database of up to 30 terabytes of uncompressed data (128 TB if only a small proportion of the database is used). We provide both, scaling experiments and runtime comparisons to previous work and the current TPC-H record holder. The main contributions of this work are: • A technique to find a better partitioning of the tables in a database to allow the execution of joins without communication effort • An algorithm that selects the first k tuples of the result set of a query with a communication effort independent from the size of the database, given certain conditions of the partitioning • An analysis of the communication effort of a delayed join that can’t be evaluated locally on a node, in comparison to the communication effort when executing the join early • The application of our algorithms to solve complex queries of the TPC-H benchmark that can’t be executed without a high amount of communication effort • The implementation of the queries in a prototype and evaluation of our algorithms on a large cluster consisting of 128 nodes for a database with up to 30 terabytes of uncompressed data (or 128 TB if only a small proportion of the database is used)