Ernestas Poškus
About Ernestas Poškus
Ernestas Poškus is an author and software engineer known for his contributions to Elasticsearch optimization and infrastructure improvements at Vinted.
Ernestas Poškus: Author
Ernestas Poškus is an established author who has contributed significantly to the field of Elasticsearch. His expertise is reflected in various technical initiatives and detailed writings that provide valuable insights into Elasticsearch optimization and resource management. Through his written work, including blog posts detailing technical evolutions, he has shared extensive knowledge and practical solutions with a broader audience.
Elasticsearch Exporter Development
Ernestas Poškus played a pivotal role in the development of a new Elasticsearch exporter written in Rust. This initiative aimed to enhance the performance and efficiency of Elasticsearch metrics collection. By leveraging the capabilities of Rust, a language known for its performance and safety, the new exporter provides a robust solution for Elasticsearch monitoring and metrics collection.
Blog Post on Elasticsearch Metrics Collection at Vinted
Ernestas Poškus authored a detailed blog post that chronicles the evolution of Elasticsearch metrics collection at Vinted. This post serves as an in-depth look at the challenges faced, solutions implemented, and the overall journey of improving Elasticsearch metrics collection within the company. It provides valuable insights and practical recommendations for others facing similar challenges.
Sidekiq Implementation for Improved Job Processing
In his role, Ernestas Poškus led the initiative to replace Resque with Sidekiq, a change that significantly improved job processing speed and resource utilization. This transition allowed for more efficient background job processing within the system, making better use of available resources and enhancing overall system performance.
Server Accounting Automation
Ernestas Poškus participated in the automation of server accounting, a project aimed at identifying discrepancies and optimizing hardware costs. This initiative involved the deployment of automated tools to better manage server resources, ensuring accurate accounting and more efficient use of hardware investments.
NUMA Resource Usage in Elasticsearch
Exploring the use of Non-Uniform Memory Access (NUMA) for efficient resource usage in Elasticsearch, Ernestas Poškus was able to configure clusters more efficiently. This exploration and subsequent implementation led to improved performance and resource management within Elasticsearch clusters.
Optimizing Elasticsearch Shard Allocation and Distribution
Ernestas Poškus shared his insights on optimizing Elasticsearch shard allocation and distribution, contributing to better resource utilization. His work in this area helps ensure that Elasticsearch operates smoothly and efficiently by making the best use of available resources.
Elasticsearch Index Settings Sensu Check
To automate the process of maintaining optimal Elasticsearch shard distribution, Ernestas Poškus developed a Sensu check. This tool advises on changes to Elasticsearch index settings, helping to automate and streamline the maintenance of shard distribution, ultimately leading to more efficient cluster configuration.