Hdconfigor: Automatically Tuning High Dimensional Configuration Parameters for Log Search Engines

Abstract

Search engines are nowadays widely applied to store and analyze logs generated by largescale distributed systems. To adapt to various workload scenarios, log search engines such as Elasticsearch usually expose a large number of performance-related configuration parameters. As manual configuring is time consuming and labor intensive, automatically tuning configuration parameters to optimize performance has been an urgent need. However, it is challenging because: 1) Due to the complex implementation, the relationship between performance and configuration parameters is difficult to model and thus the objective function is actually a black box; 2) In addition to application parameters, JVM and kernel parameters are also closely related to the performance and together they construct a high dimensional configuration space; 3) To iteratively search for the best configuration, a tool is necessary to automatically deploy the newly generated configuration and launch tests to measure the corresponding performance. To address these challenges, this paper designs and implements HDConfigor, an automatic holistic configuration parameter tuning tool for log search engines. In order to solve the high dimensional optimization problem, we propose a modified Random EMbedding Bayesian Optimization algorithm (mREMBO) in HDConfigor which is a black-box approach. Instead of directly using a black-box optimization algorithm such as Bayesian optimization (BO), mREMBO first generates a lower dimensional embedded space through introducing a random embedding matrix and then performs BO in this embedded space. Therefore, HDConfigor is able to find a competitive configuration automatically and quickly. We evaluate HDConfigor in an Elasticsearch cluster with different workload scenarios. Experimental results show that compared with the default configuration, the best relative median indexing results achieved by mREMBO can reach 2.07×. In addition, under the same number of trials, mREMBO is able to find a configuration with at least a further 10.31% improvement in throughput compared to Random search, Simulated Annealing and BO. INDEX TERMS Log search engine, configuration parameter tuning, black-box optimization, Bayesian optimization, random embedding.

Automatic configuration of the Cassandra database using irace

“…These methods, however, are very complex and not easy to implement. Dou, Chen & Zheng (2020) model the database configuration problem as a black-box problem that they solve using mREMBO (modified Random EMbedding Bayesian Optimization) through the generation of an embedded space of smallest dimension using a random embedding matrix. This work configures the full stack, namely the OS kernel, the JVM and a set of Elasticsearch parameters.…”

Tuning of Elasticsearch Configuration: Parameter Optimization Through Simultaneous Perturbation Stochastic Approximation

“…In Dou et al ( 2020 ), the authors present HDConfigor , an automatic full-stack configuration parameter tuning tool for log search engines. HDConfigor solves the high dimensional black-box optimization problem by using an algorithm that introduces a random embedding matrix to generate an embedded space.…”

JointConf: Jointly autotuning configuration parameters for modularized graph databases

“…For instance, JanusGraph itself mainly focuses on graph serialization and query execution, while providing adapters to integrate third‐party softwares as its functional module for data storage and indices. Unfortunately, although there are already a few significant works toward automatically tuning parameters for different databases such as HBase, 4 Elasticsearch, 5 RocksDB, 6 and MySQL, 7 these solutions cannot be directly applied in the scenarios of modularized GDBs because they solely consider one specific software. What is worse, due to the complicated interactions across different modules, sequentially tuning each software with previous solutions may also fail to efficiently find the optimal configuration for modularized GDBs.…”

“…Broadly speaking, prior approaches can be divided into learning based 4,9,12–14 and search based 5,15–17 . Unfortunately, as described in Section 2, none of them can simultaneously address all these challenges.…”

“…To evaluate the effectiveness of JointConf , we select three frequently used graph query benchmark applications and compare BO_dropout with four other widely used black‐box optimization algorithms: Random search , 8 BestConfig , 16 Bayesian optimization , 19 and REMBO 5 . Experimental results on a local distributed JanusGraph cluster show that for the k ‐hop neighborhood benchmark, BO_dropout is able to further reduce the query time by at least 5.66% compared with the baselines under the same observation times constraint.…”